analise-de-dados-financeiros-no-pandas.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introdução\n",
    "\n",
    "Aqui veremos uma análise de dados financeiros e de performance de uma empresa fictícia, para mostrar as capacidades de análise e visualização de dados no Python, através dos pacotes [pandas](https://pandas.pydata.org/) e [matplotlib](https://matplotlib.org/). A empresa possui os seguintes dados:\n",
    "\n",
    "- *Visitas*: Base de quantas visitas temos no site, separadas por departamento, produto e região do país\n",
    "- *Vendas*: Base de quantas vendas tivemos, separadas por departamento, produto e região do país\n",
    "- *Financeiro*: Dados financeiros da venda, como receita com preço (após descontos), receita com frete (após descontos), custo do produto, custo de entrega, gastos com marketing e descontos dados em preço e frete.\n",
    "\n",
    "Propõe-se uma análise dos indicadores de performance da empresa, considerando as informações disponíveis. Vamos importar os pacotes necessários e os dados:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "fin = pd.read_csv('~/Downloads/Financeiro.csv', sep=';', encoding='windows-1252', decimal=b',')\n",
    "ven = pd.read_csv('~/Downloads/Vendas.csv', sep=';', encoding='windows-1252', decimal=b',')\n",
    "vis = pd.read_csv('~/Downloads/visitas.csv', sep=';', encoding='windows-1252', decimal=b',')\n",
    "\n",
    "df = pd.merge(fin, ven)\n",
    "df = pd.merge(df, vis)\n",
    "df.Mes = pd.to_datetime(df.Mes)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Uma amostra da base de dados inicial, com dados do setor financeiro, de vendas e visitas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Mes</th>\n",
       "      <th>Departamento</th>\n",
       "      <th>UF</th>\n",
       "      <th>Item</th>\n",
       "      <th>Faturamento_Produto</th>\n",
       "      <th>Faturamento_Frete</th>\n",
       "      <th>Custo_Produto</th>\n",
       "      <th>Custo_Frete</th>\n",
       "      <th>Custo_Mkt</th>\n",
       "      <th>Desconto_Produto</th>\n",
       "      <th>Desconto_Frete</th>\n",
       "      <th>Vendas</th>\n",
       "      <th>Visitas</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2018-06-01</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>SP</td>\n",
       "      <td>1</td>\n",
       "      <td>149250.0</td>\n",
       "      <td>42984.0</td>\n",
       "      <td>71640</td>\n",
       "      <td>35820</td>\n",
       "      <td>5970</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4776.0</td>\n",
       "      <td>590</td>\n",
       "      <td>14304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2018-06-01</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>SP</td>\n",
       "      <td>2</td>\n",
       "      <td>148050.0</td>\n",
       "      <td>30456.0</td>\n",
       "      <td>76140</td>\n",
       "      <td>25380</td>\n",
       "      <td>4230</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3384.0</td>\n",
       "      <td>439</td>\n",
       "      <td>14258</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2018-06-01</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>SP</td>\n",
       "      <td>3</td>\n",
       "      <td>133650.0</td>\n",
       "      <td>21384.0</td>\n",
       "      <td>89100</td>\n",
       "      <td>17820</td>\n",
       "      <td>2970</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2376.0</td>\n",
       "      <td>292</td>\n",
       "      <td>14325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2018-07-01</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>SP</td>\n",
       "      <td>1</td>\n",
       "      <td>314712.5</td>\n",
       "      <td>94784.0</td>\n",
       "      <td>177720</td>\n",
       "      <td>88860</td>\n",
       "      <td>74050</td>\n",
       "      <td>55537.5</td>\n",
       "      <td>23696.0</td>\n",
       "      <td>1469</td>\n",
       "      <td>14873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2018-07-01</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>SP</td>\n",
       "      <td>2</td>\n",
       "      <td>392105.0</td>\n",
       "      <td>84352.0</td>\n",
       "      <td>237240</td>\n",
       "      <td>79080</td>\n",
       "      <td>65900</td>\n",
       "      <td>69195.0</td>\n",
       "      <td>21088.0</td>\n",
       "      <td>1333</td>\n",
       "      <td>14849</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Mes Departamento  UF  Item  Faturamento_Produto  Faturamento_Frete  \\\n",
       "0 2018-06-01     Cadeiras  SP     1             149250.0            42984.0   \n",
       "1 2018-06-01     Cadeiras  SP     2             148050.0            30456.0   \n",
       "2 2018-06-01     Cadeiras  SP     3             133650.0            21384.0   \n",
       "3 2018-07-01     Cadeiras  SP     1             314712.5            94784.0   \n",
       "4 2018-07-01     Cadeiras  SP     2             392105.0            84352.0   \n",
       "\n",
       "   Custo_Produto  Custo_Frete  Custo_Mkt  Desconto_Produto  Desconto_Frete  \\\n",
       "0          71640        35820       5970               0.0          4776.0   \n",
       "1          76140        25380       4230               0.0          3384.0   \n",
       "2          89100        17820       2970               0.0          2376.0   \n",
       "3         177720        88860      74050           55537.5         23696.0   \n",
       "4         237240        79080      65900           69195.0         21088.0   \n",
       "\n",
       "   Vendas  Visitas  \n",
       "0     590    14304  \n",
       "1     439    14258  \n",
       "2     292    14325  \n",
       "3    1469    14873  \n",
       "4    1333    14849  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora, vamos calcular faturamentos, custos e descontos totais, bem como os lucros de produto, frete e o lucro líquido:\n",
    "\n",
    "$F_{bruto} = F_{produto} + F_{frete}$ (faturamentos)\n",
    "\n",
    "$C_{total} = C_{produto} + C_{frete} + C_{mkt}$ (custos)\n",
    "\n",
    "$D_{total} = D_{produto} + D_{frete}$ (descontos)\n",
    "\n",
    "Os lucros:\n",
    "\n",
    "$L_{produto} = F_{produto} - C_{produto} - D_{produto}$ (produto)\n",
    "\n",
    "$L_{frete} = F_{frete} - C_{frete} - D_{frete}$ (fretes)\n",
    "\n",
    "$L_{liquido} = F_{bruto} - C_{total} - D_{total}$ (líquido)\n",
    "\n",
    "A seguir, calcula-se o custo marginal (o custo de se ofertar uma unidade a mais) dos itens pela relação:\n",
    "\n",
    "$C_{marginal} = \\frac{C_{total}}{\\textrm{Vendas}}$\n",
    "\n",
    "Podemos inferir o preço médio praticado, através da razão entre o faturamento bruto e as quantidades vendidas:\n",
    "\n",
    "$P_{médio} = \\frac{F_{bruto}}{\\textrm{Vendas}}$\n",
    "\n",
    "Agora, sabendo o custo marginal e o preço dos itens, podemos calcular as margens de lucro (_mark-ups_) em porcentagem:\n",
    "\n",
    "$M = \\frac{P_{médio}-C_{marginal}}{P_{médio}} * 100$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['Faturamento_Bruto'] = df.Faturamento_Produto + df.Faturamento_Frete\n",
    "df['Custo_Total']       = df.Custo_Produto + df.Custo_Frete + df.Custo_Mkt\n",
    "df['Desconto_Total']    = df.Desconto_Frete + df.Desconto_Produto\n",
    "\n",
    "df['Lucro_Produto'] = df.Faturamento_Produto - df.Custo_Produto - df.Desconto_Produto\n",
    "df['Lucro_Frete']   = df.Faturamento_Frete - df.Custo_Frete - df.Desconto_Frete\n",
    "df['Lucro_Liquido'] = df.Faturamento_Bruto - df.Custo_Total - df.Desconto_Total\n",
    "\n",
    "df['Custo_Marginal'] = (df.Custo_Total) / df.Vendas\n",
    "df['Preco_Medio']    = df.Faturamento_Bruto / df.Vendas\n",
    "df['Markup']      = (df.Preco_Medio - df.Custo_Marginal)/df.Preco_Medio * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Análise da performance da empresa\n",
    "\n",
    "## Faturamento e lucro\n",
    "\n",
    "Inicialmente, vamos ver como foi a performance da empresa ao longo do período estudado (dados em milhares de R$):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Faturamento_Bruto</th>\n",
       "      <th>Custo_Total</th>\n",
       "      <th>Desconto_Total</th>\n",
       "      <th>Lucro_Liquido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-06-01</th>\n",
       "      <td>3861.4535</td>\n",
       "      <td>3162.896</td>\n",
       "      <td>146.8465</td>\n",
       "      <td>551.7110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-01</th>\n",
       "      <td>9071.9396</td>\n",
       "      <td>8522.998</td>\n",
       "      <td>1483.1344</td>\n",
       "      <td>-934.1928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-08-01</th>\n",
       "      <td>7561.4144</td>\n",
       "      <td>6807.553</td>\n",
       "      <td>867.2816</td>\n",
       "      <td>-113.4202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-01</th>\n",
       "      <td>4937.5492</td>\n",
       "      <td>4093.750</td>\n",
       "      <td>176.5648</td>\n",
       "      <td>667.2344</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Faturamento_Bruto  Custo_Total  Desconto_Total  Lucro_Liquido\n",
       "Mes                                                                      \n",
       "2018-06-01          3861.4535     3162.896        146.8465       551.7110\n",
       "2018-07-01          9071.9396     8522.998       1483.1344      -934.1928\n",
       "2018-08-01          7561.4144     6807.553        867.2816      -113.4202\n",
       "2018-09-01          4937.5492     4093.750        176.5648       667.2344"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby('Mes')['Faturamento_Bruto','Custo_Total','Desconto_Total','Lucro_Liquido'].agg('sum', margins=True)/1000\n",
    "temp"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f0753aa780>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp.plot(title='Resultados da empresa (R$ mil)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Em junho, a empresa apresentou lucro líquido de R\\\\$ 551 mil, seguido de prejuízos em julho (-R\\\\$ 934 mil) e agosto (-R\\\\$ 113 mil), para uma recuperação no mês de setembro (R\\\\$ 667 mil). O resultado agregado no período foi positivo (R\\\\$ 171 mil).\n",
    "\n",
    "Agora, vamos ver quais são os departamentos com maior faturamento e lucro, no somatório dos meses observados e dos produtos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Faturamento_Bruto</th>\n",
       "      <th>Lucro_Liquido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <td>464448.7</td>\n",
       "      <td>111036.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cadeiras</th>\n",
       "      <td>6141458.0</td>\n",
       "      <td>-234484.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "      <td>18826450.0</td>\n",
       "      <td>294780.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       Faturamento_Bruto  Lucro_Liquido\n",
       "Departamento                                           \n",
       "Bolas de Gude                   464448.7       111036.4\n",
       "Cadeiras                       6141458.0      -234484.0\n",
       "Maquinas Fotogr ficas         18826450.0       294780.0"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(['Departamento'])['Faturamento_Bruto','Lucro_Liquido'].sum().round(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vemos que, ao longo dos meses observados, o departamento de máquinas fotográficas obteve a maior rentabilidade, considerando todas as regiões e itens ofertados. Nota-se também que o setor de cadeiras apresentou prejuízos.\n",
    "\n",
    "Agora, a evolução da performance mês a mês, na soma dos produtos:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">Faturamento_Bruto</th>\n",
       "      <th colspan=\"3\" halign=\"left\">Lucro_Liquido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <th>Cadeiras</th>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <th>Cadeiras</th>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-06-01</th>\n",
       "      <td>84315.5</td>\n",
       "      <td>868238.0</td>\n",
       "      <td>2908900.0</td>\n",
       "      <td>28395.0</td>\n",
       "      <td>279796.0</td>\n",
       "      <td>243520.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-01</th>\n",
       "      <td>155948.6</td>\n",
       "      <td>2431401.0</td>\n",
       "      <td>6484590.0</td>\n",
       "      <td>38980.2</td>\n",
       "      <td>-214633.0</td>\n",
       "      <td>-758540.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-08-01</th>\n",
       "      <td>130574.4</td>\n",
       "      <td>1792560.0</td>\n",
       "      <td>5638280.0</td>\n",
       "      <td>32044.8</td>\n",
       "      <td>-329945.0</td>\n",
       "      <td>184480.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-01</th>\n",
       "      <td>93610.2</td>\n",
       "      <td>1049259.0</td>\n",
       "      <td>3794680.0</td>\n",
       "      <td>11616.4</td>\n",
       "      <td>30298.0</td>\n",
       "      <td>625320.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Faturamento_Bruto                                  Lucro_Liquido  \\\n",
       "Departamento     Bolas de Gude   Cadeiras Maquinas Fotogr ficas Bolas de Gude   \n",
       "Mes                                                                             \n",
       "2018-06-01             84315.5   868238.0             2908900.0       28395.0   \n",
       "2018-07-01            155948.6  2431401.0             6484590.0       38980.2   \n",
       "2018-08-01            130574.4  1792560.0             5638280.0       32044.8   \n",
       "2018-09-01             93610.2  1049259.0             3794680.0       11616.4   \n",
       "\n",
       "                                              \n",
       "Departamento  Cadeiras Maquinas Fotogr ficas  \n",
       "Mes                                           \n",
       "2018-06-01    279796.0              243520.0  \n",
       "2018-07-01   -214633.0             -758540.0  \n",
       "2018-08-01   -329945.0              184480.0  \n",
       "2018-09-01     30298.0              625320.0  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['Mes','Departamento'])['Faturamento_Bruto','Lucro_Liquido'].sum().round(2)\n",
    "temp.unstack()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aqui é possível ver uma tendência do mercado em geral: percebe-se um aquecimento das vendas nos meses de julho e agosto. O departamento de bolas de gude apresentou um crescimento expressivo de junho para julho. Para ter uma melhor ideia da evolução dos indicadores, vejamos os dados em termos da variação em relação ao mês anterior (perde-se uma observação, pois não dispomos dos dados de maio):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"3\" halign=\"left\">Faturamento_Bruto</th>\n",
       "      <th colspan=\"3\" halign=\"left\">Lucro_Liquido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <th>Cadeiras</th>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <th>Cadeiras</th>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-06-01</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-01</th>\n",
       "      <td>84.96</td>\n",
       "      <td>180.04</td>\n",
       "      <td>122.92</td>\n",
       "      <td>37.28</td>\n",
       "      <td>-176.71</td>\n",
       "      <td>-411.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-08-01</th>\n",
       "      <td>-16.27</td>\n",
       "      <td>-26.27</td>\n",
       "      <td>-13.05</td>\n",
       "      <td>-17.79</td>\n",
       "      <td>53.73</td>\n",
       "      <td>-124.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-01</th>\n",
       "      <td>-28.31</td>\n",
       "      <td>-41.47</td>\n",
       "      <td>-32.70</td>\n",
       "      <td>-63.75</td>\n",
       "      <td>-109.18</td>\n",
       "      <td>238.96</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Faturamento_Bruto                                Lucro_Liquido  \\\n",
       "Departamento     Bolas de Gude Cadeiras Maquinas Fotogr ficas Bolas de Gude   \n",
       "Mes                                                                           \n",
       "2018-06-01                 NaN      NaN                   NaN           NaN   \n",
       "2018-07-01               84.96   180.04                122.92         37.28   \n",
       "2018-08-01              -16.27   -26.27                -13.05        -17.79   \n",
       "2018-09-01              -28.31   -41.47                -32.70        -63.75   \n",
       "\n",
       "                                             \n",
       "Departamento Cadeiras Maquinas Fotogr ficas  \n",
       "Mes                                          \n",
       "2018-06-01        NaN                   NaN  \n",
       "2018-07-01    -176.71               -411.49  \n",
       "2018-08-01      53.73               -124.32  \n",
       "2018-09-01    -109.18                238.96  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp2 = temp.unstack().pct_change()*100\n",
    "temp2.round(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aqui percebem-se mais claramente as tendências de receitas. Para o faturamento bruto, observa-se um salto na variação de junho para julho, seguido de queda nos meses seguintes. Aqui, os setores de cadeiras e máquinas fotográficas merecem destaque, pois em alguns meses existe uma relação inversa entre o faturamento bruto e o lucro líquido dos departamentos. Isto sugere que precisamos analisar cuidadosamente a estrutura de custos e descontos destes setores, o que será feito adiante.\n",
    "\n",
    "A seguir, veremos no gráfico a evolução desses indicadores:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f0757306a0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp2['Faturamento_Bruto'].plot(title='Faturamento Bruto (var. %)')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f07579cd30>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp2['Lucro_Liquido'].plot(title='Lucro Líquido (var. %)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vamos analisar agora quais os itens mais rentáveis, em termos de lucro total ao longo do período:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Faturamento_Bruto</th>\n",
       "      <th>Lucro_Liquido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>170946.35</td>\n",
       "      <td>40743.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>154184.50</td>\n",
       "      <td>37215.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>139317.85</td>\n",
       "      <td>33077.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>1946127.50</td>\n",
       "      <td>16035.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2102926.50</td>\n",
       "      <td>22143.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2092404.00</td>\n",
       "      <td>-272662.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>4103445.00</td>\n",
       "      <td>-244990.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5514950.00</td>\n",
       "      <td>22120.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9208055.00</td>\n",
       "      <td>517650.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Faturamento_Bruto  Lucro_Liquido\n",
       "Departamento          Item                                  \n",
       "Bolas de Gude         1             170946.35        40743.7\n",
       "                      2             154184.50        37215.0\n",
       "                      3             139317.85        33077.7\n",
       "Cadeiras              1            1946127.50        16035.0\n",
       "                      2            2102926.50        22143.0\n",
       "                      3            2092404.00      -272662.0\n",
       "Maquinas Fotogr ficas 1            4103445.00      -244990.0\n",
       "                      2            5514950.00        22120.0\n",
       "                      3            9208055.00       517650.0"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['Departamento','Item'])['Faturamento_Bruto','Lucro_Liquido'].sum().round(2)\n",
    "temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "No total dos meses observados, os itens vendidos mais lucrativos foram:\n",
    "\n",
    "- Bolas de gude: item 1\n",
    "- Cadeiras: item 2\n",
    "- Máquinas fotográficas: item 3\n",
    "\n",
    "Agora, vamos ver no gráfico como andam as margens de lucro mínimas, médias e máximas para todo o período, de cada item:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>min</th>\n",
       "      <th>mean</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>4.76</td>\n",
       "      <td>26.96</td>\n",
       "      <td>41.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4.76</td>\n",
       "      <td>26.96</td>\n",
       "      <td>41.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.76</td>\n",
       "      <td>26.96</td>\n",
       "      <td>41.02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>3.77</td>\n",
       "      <td>18.87</td>\n",
       "      <td>40.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.26</td>\n",
       "      <td>20.11</td>\n",
       "      <td>40.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-8.75</td>\n",
       "      <td>7.27</td>\n",
       "      <td>29.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>-11.11</td>\n",
       "      <td>2.95</td>\n",
       "      <td>14.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-13.45</td>\n",
       "      <td>8.97</td>\n",
       "      <td>23.42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3.51</td>\n",
       "      <td>13.53</td>\n",
       "      <td>21.90</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                              min   mean    max\n",
       "Departamento          Item                     \n",
       "Bolas de Gude         1      4.76  26.96  41.02\n",
       "                      2      4.76  26.96  41.02\n",
       "                      3      4.76  26.96  41.02\n",
       "Cadeiras              1      3.77  18.87  40.99\n",
       "                      2      5.26  20.11  40.76\n",
       "                      3     -8.75   7.27  29.12\n",
       "Maquinas Fotogr ficas 1    -11.11   2.95  14.67\n",
       "                      2    -13.45   8.97  23.42\n",
       "                      3      3.51  13.53  21.90"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['Departamento','Item'])['Markup'].agg(['min','mean','max']).round(2)\n",
    "temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calculando as margens médias, descobrimos que o departamento de bolas de gude exerce a mesma margem de lucro para todos os seus itens, que por sinal são as maiores dentro da empresa (aprox. 27%), embora tenha aplicado margens de até 41% sobre os itens.\n",
    "\n",
    "O departamento de cadeiras exerce margens um pouco menores (entre 7% e 18%, na média), embora também tenha aplicado margens de quase 40% (itens 1 e 2). No caso do item 3, acende-se um sinal de alerta, com margens mínimas de -8%, ou seja, itens vendidos com preços vendidos abaixo dos custos.\n",
    "\n",
    "O mesmo aconteceu no departamento de máquinas fotográficas para os itens 1 e 2. Em geral, as margens médias deste departamento são um pouco menores do que nos departamentos de bolas de gude e cadeiras. Anteriormente vimos a capacidade de faturamento deste departamento, portanto aqui percebe-se um potencial para aumento de preços, visando a saúde financeira da empresa.\n",
    "\n",
    "Vamos ver a evolução dos dados acima no tempo, para as médias, no gráfico abaixo:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f075828390>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.groupby(['Mes','Departamento'])['Markup'].agg(['mean']).unstack().plot(title='Margens médias de lucro (%)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Estrutura de custos\n",
    "\n",
    "Estudar a estrutura de custos é vital para buscar novas estratégias de aumentar a rentabilidade da empresa: enquanto estratégias pelo lado da receita dependem da demanda dos consumidores, alguns cortes de custos podem ser realizadas (i) no curto prazo e (ii) dentro dos limites operacionais da empresa.\n",
    "\n",
    "Primeiramente, vamos ver quais são os custos médios observados ao longo dos meses, de produto, frete e marketing, por departamento e item:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f07588f9e8>"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp = df.groupby('Mes')['Custo_Produto','Custo_Frete','Custo_Mkt','Custo_Total']\n",
    "temp.mean().plot(kind='line')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Custo_Produto</th>\n",
       "      <th>Custo_Frete</th>\n",
       "      <th>Custo_Mkt</th>\n",
       "      <th>Custo_Total</th>\n",
       "      <th>Custo_Marginal</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>1628.33</td>\n",
       "      <td>7013.58</td>\n",
       "      <td>1497.42</td>\n",
       "      <td>10139.33</td>\n",
       "      <td>12.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1445.08</td>\n",
       "      <td>6299.92</td>\n",
       "      <td>1365.00</td>\n",
       "      <td>9110.00</td>\n",
       "      <td>11.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1330.75</td>\n",
       "      <td>5723.25</td>\n",
       "      <td>1222.25</td>\n",
       "      <td>8276.25</td>\n",
       "      <td>12.19</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>69860.00</td>\n",
       "      <td>39945.00</td>\n",
       "      <td>23593.33</td>\n",
       "      <td>133398.33</td>\n",
       "      <td>226.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>88200.00</td>\n",
       "      <td>33567.50</td>\n",
       "      <td>20380.00</td>\n",
       "      <td>142147.50</td>\n",
       "      <td>282.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>120000.00</td>\n",
       "      <td>27325.00</td>\n",
       "      <td>17193.33</td>\n",
       "      <td>164518.33</td>\n",
       "      <td>403.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>285208.33</td>\n",
       "      <td>21691.67</td>\n",
       "      <td>22845.00</td>\n",
       "      <td>329745.00</td>\n",
       "      <td>1300.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>376250.00</td>\n",
       "      <td>20070.00</td>\n",
       "      <td>21435.00</td>\n",
       "      <td>417755.00</td>\n",
       "      <td>1785.14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>632458.33</td>\n",
       "      <td>16920.83</td>\n",
       "      <td>17797.50</td>\n",
       "      <td>667176.67</td>\n",
       "      <td>3279.55</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Custo_Produto  Custo_Frete  Custo_Mkt  \\\n",
       "Departamento          Item                                          \n",
       "Bolas de Gude         1           1628.33      7013.58    1497.42   \n",
       "                      2           1445.08      6299.92    1365.00   \n",
       "                      3           1330.75      5723.25    1222.25   \n",
       "Cadeiras              1          69860.00     39945.00   23593.33   \n",
       "                      2          88200.00     33567.50   20380.00   \n",
       "                      3         120000.00     27325.00   17193.33   \n",
       "Maquinas Fotogr ficas 1         285208.33     21691.67   22845.00   \n",
       "                      2         376250.00     20070.00   21435.00   \n",
       "                      3         632458.33     16920.83   17797.50   \n",
       "\n",
       "                            Custo_Total  Custo_Marginal  \n",
       "Departamento          Item                               \n",
       "Bolas de Gude         1        10139.33           12.09  \n",
       "                      2         9110.00           11.47  \n",
       "                      3         8276.25           12.19  \n",
       "Cadeiras              1       133398.33          226.70  \n",
       "                      2       142147.50          282.74  \n",
       "                      3       164518.33          403.79  \n",
       "Maquinas Fotogr ficas 1       329745.00         1300.18  \n",
       "                      2       417755.00         1785.14  \n",
       "                      3       667176.67         3279.55  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['Departamento','Item'])['Custo_Produto','Custo_Frete','Custo_Mkt','Custo_Total','Custo_Marginal'].mean().round(2)\n",
    "temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Os custos, em geral, são função da quantidade vendida de itens. O custo marginal, como exposto anteriormente, é uma métrica de custos que é ponderada pelas vendas. No departamento de bolas de gude, o item 3 foi o que apresentou o maior custo marginal. No departamento de cadeiras, o item 3 é o que apresentou o maior custo, e no departamento de máquinas fotográficas o item 3. Quaisquer estratégias de aumento de rentabilidade via redução de custos deve passar pelo estudo destes itens.\n",
    "\n",
    "Todavia, nem sempre é possível fazer grandes intervenções nas negociações com fornecedores e custos de transporte: é preciso saber como cada etapa contribui sobre os custos totais. Vamos analisar em cada item a esta participação média de cada etapa dos custos sobre o total, em porcentagens:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th>Produto pelo Total</th>\n",
       "      <th>Frete pelo Total</th>\n",
       "      <th>Mkt pelo Total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>1</td>\n",
       "      <td>16.060</td>\n",
       "      <td>69.172</td>\n",
       "      <td>14.768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>2</td>\n",
       "      <td>15.863</td>\n",
       "      <td>69.154</td>\n",
       "      <td>14.984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>3</td>\n",
       "      <td>16.079</td>\n",
       "      <td>69.153</td>\n",
       "      <td>14.768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>1</td>\n",
       "      <td>52.369</td>\n",
       "      <td>29.944</td>\n",
       "      <td>17.686</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>2</td>\n",
       "      <td>62.048</td>\n",
       "      <td>23.615</td>\n",
       "      <td>14.337</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>3</td>\n",
       "      <td>72.940</td>\n",
       "      <td>16.609</td>\n",
       "      <td>10.451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>1</td>\n",
       "      <td>86.494</td>\n",
       "      <td>6.578</td>\n",
       "      <td>6.928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>2</td>\n",
       "      <td>90.065</td>\n",
       "      <td>4.804</td>\n",
       "      <td>5.131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>3</td>\n",
       "      <td>94.796</td>\n",
       "      <td>2.536</td>\n",
       "      <td>2.668</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Departamento  Item  Produto pelo Total  Frete pelo Total  \\\n",
       "0          Bolas de Gude     1              16.060            69.172   \n",
       "1          Bolas de Gude     2              15.863            69.154   \n",
       "2          Bolas de Gude     3              16.079            69.153   \n",
       "3               Cadeiras     1              52.369            29.944   \n",
       "4               Cadeiras     2              62.048            23.615   \n",
       "5               Cadeiras     3              72.940            16.609   \n",
       "6  Maquinas Fotogr ficas     1              86.494             6.578   \n",
       "7  Maquinas Fotogr ficas     2              90.065             4.804   \n",
       "8  Maquinas Fotogr ficas     3              94.796             2.536   \n",
       "\n",
       "   Mkt pelo Total  \n",
       "0          14.768  \n",
       "1          14.984  \n",
       "2          14.768  \n",
       "3          17.686  \n",
       "4          14.337  \n",
       "5          10.451  \n",
       "6           6.928  \n",
       "7           5.131  \n",
       "8           2.668  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = temp.reset_index()\n",
    "temp['Produto pelo Total'] = temp.Custo_Produto / temp.Custo_Total * 100\n",
    "temp['Frete pelo Total'] = temp.Custo_Frete / temp.Custo_Total * 100\n",
    "temp['Mkt pelo Total'] = temp.Custo_Mkt / temp.Custo_Total * 100\n",
    "temp.drop(['Custo_Produto','Custo_Frete','Custo_Mkt','Custo_Total','Custo_Marginal'], axis=1, inplace=True)\n",
    "temp.round(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vimos anteriormente que no departamento de bolas de gude o item 3 possui os maiores custos marginais de comercialização. Dentro deste departamento como um todo, as estruturas de custo são similares, e o custo do frete é o principal fator (cerca de 69% do custo total). Tratando-se de um produto simples e barato, estratégias de otimização da logística podem trazer grandes contribuições para a rentabilidade do departamento como um todo.\n",
    "\n",
    "No departamento de cadeiras, a distribuição é mais equilibrada e heterogênea: os custos de produto representam cerca de 60% dos custos totais. No caso do item 1, observam-se a maior fatia dos custos com marketing (aprox. 17%). A aplicação bem-sucedida dos dispêndios com marketing depende da propensão à vender dadas as visitas no site, ponto que será visto adiante.\n",
    "\n",
    "No departamento de máquinas fotográficas, os custos de produto representam cerca de 90% dos custos totais, e os dispêndios com frete e marketing são baixos (cerca de 5% cada). Analisando o faturamento da nossa empresa, vimos que os itens do departamento de máquinas fotográficas são altamente lucrativos e de alto valor agregado. Aqui, negociações com fornecedores que reduzam o custo marginal dos itens trarão grandes benefícios para a empresa."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Descontos, visitas e vendas\n",
    "\n",
    "Os descontos representam uma peça-chave da estrutura de custos da empresa, pois com eles pode-se obter uma vantagem competitiva no mercado. Portanto, a análise aqui concentra-se não somente em verificar qual a contribuição dos descontos para a estrutura de custos totais da empresa, mas em como os descontos têm alavancado as vendas."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th>Desconto_Produto</th>\n",
       "      <th>Desconto_Frete</th>\n",
       "      <th>Desconto_Total</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>1</td>\n",
       "      <td>363.97</td>\n",
       "      <td>346.92</td>\n",
       "      <td>710.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>2</td>\n",
       "      <td>333.93</td>\n",
       "      <td>303.53</td>\n",
       "      <td>637.46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>3</td>\n",
       "      <td>303.89</td>\n",
       "      <td>273.21</td>\n",
       "      <td>577.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>1</td>\n",
       "      <td>22079.88</td>\n",
       "      <td>5362.83</td>\n",
       "      <td>27442.71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>2</td>\n",
       "      <td>26719.62</td>\n",
       "      <td>4531.50</td>\n",
       "      <td>31251.12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>3</td>\n",
       "      <td>28743.50</td>\n",
       "      <td>3827.00</td>\n",
       "      <td>32570.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>1</td>\n",
       "      <td>16527.92</td>\n",
       "      <td>16096.67</td>\n",
       "      <td>32624.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>2</td>\n",
       "      <td>24069.17</td>\n",
       "      <td>15911.67</td>\n",
       "      <td>39980.83</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>3</td>\n",
       "      <td>43622.92</td>\n",
       "      <td>13400.83</td>\n",
       "      <td>57023.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Departamento  Item  Desconto_Produto  Desconto_Frete  \\\n",
       "0          Bolas de Gude     1            363.97          346.92   \n",
       "1          Bolas de Gude     2            333.93          303.53   \n",
       "2          Bolas de Gude     3            303.89          273.21   \n",
       "3               Cadeiras     1          22079.88         5362.83   \n",
       "4               Cadeiras     2          26719.62         4531.50   \n",
       "5               Cadeiras     3          28743.50         3827.00   \n",
       "6  Maquinas Fotogr ficas     1          16527.92        16096.67   \n",
       "7  Maquinas Fotogr ficas     2          24069.17        15911.67   \n",
       "8  Maquinas Fotogr ficas     3          43622.92        13400.83   \n",
       "\n",
       "   Desconto_Total  \n",
       "0          710.89  \n",
       "1          637.46  \n",
       "2          577.10  \n",
       "3        27442.71  \n",
       "4        31251.12  \n",
       "5        32570.50  \n",
       "6        32624.58  \n",
       "7        39980.83  \n",
       "8        57023.75  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['Departamento','Item'])['Desconto_Produto','Desconto_Frete','Desconto_Total','Vendas','Preco_Medio'].mean().reset_index().round(2)\n",
    "temp.drop(['Vendas','Preco_Medio'], axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora, podemos dividir os descontos por itens pelo respectivo volume de vendas, para descobrir qual têm sido o desconto necessário para vender uma unidade adicional de algum item, descritos pelas colunas `DescProd_Vendas`, `DescFret_Vendas` e `DescTot_Vendas`, que representam os descontos de produto, frete e total divididos pelas vendas, respectivamente. É possível ainda obter os descontos no preço médio dos itens em porcentagens através da relação:\n",
    "\n",
    "$ \\textrm{Desconto no preço} = \\frac{\\textrm{DescTot_Vendas}}{P_{médio}} * 100 $"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f075923d68>"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAX4AAAGPCAYAAABbFLtAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzt3XmcZFV9/vHPww6CKMyAbMMgIgqKBEaUoD9BoiIBNQYDiAbXUSNqXBIxUSFmIxo1CiqiILgENSqKyBoBkSjLDDsigggyDrKvCsLA8/vjnpKiqe5punr6nqn7vF+venXdpe59umvmW7fOvfcc2SYiIrpjhbYDRETEzErhj4jomBT+iIiOSeGPiOiYFP6IiI5J4Y+I6JgU/oiKSFpV0s8kPWkKr91G0k+WRa4YLSn80TpJ10q6V9Ldku6Q9BNJb5VU5b9PSQdL+uoy2vx84Czbvy37erWkGyT9StLOfRk2L3+nFXvzbF8C3CFpz2WULUZElf+xopP2tL0WsClwCPB+4Mh2I7XiLcBXACStRPO32A54B3BY33qfBt5j+8Exr/9a2UbEuFL4oyq277R9PLA3sL+kZ8Afm0D+U9KvJd0o6XBJq5dlsySdUL4t3Cbpx71vC5I2kfQdSTdLulXSYWX+CpI+KOk6STdJ+rKktcuyuZIsaf+yv1sk/WNZthvwD8Deku6RdHGZv6Gk48v+r5b05t7vJGkHSQsk3VWyf2LQ7y5pDrA5cG6ZtS7wG9s3AP8LPLmst1eZf86AzZwJ7Cpp1am+BzH6UvijSrbPAxYBzy+z/gN4KrAt8BRgI+DDZdl7y7qzgfVpCrNLM8gJwHXA3PKar5fXvK48dqEpqGvyyCNqgOcBWwK7Ah+W9HTbJwP/BnzD9pq2n1XWPbZk2BDYC/g3SbuWZZ8CPmX78TSF/Zvj/NrPBK6xvaRM3wysK2lj4EXA5ZLWBD4IfGCcv9tvgAdK7oiBUvijZouBdSQJeDPwbtu32b6bpvjuU9Z7ANgA2NT2A7Z/7KYTqh1oCvHf2f6d7ftsn11esx/wCdvX2L6HppDuU5pXev7J9r22LwYuBp7FAJI2ofmQeH/Zx0XAF4HX9uV7iqRZtu8Z50gd4AnA3b0J2w8BbwO+Bbyv/A0+AhwKPFPSGZJO6X0r6nN32VbEQCn8UbONgNtojuTXABaW5pw7gJPLfICPAVcDp0q6RtKBZf4mwHV9R9D9NqT5JtBzHbASzTeGnt/2Pf89zbeCQTYEeh9I/dvbqDx/I823lZ9LOl/SHuNs53Zgrf4Ztn9o+7m2XwA8BMwDjqY5D/A64J9pPmT6rQXcMc4+Ilhp6atEzDxJz6YpnGcDtwD3AluXpoxHKAX3vcB7JW0NnCHpfOB6YI6klQYU/8U0J5J75gBLgBuBjZcSb2yXtr1vJmv1Ff85wG9KvquAfct5h1cC35K0ru3fjdnOJcCTB+Ut33oOA94JzAJWtH2dpN8C2/SttyGwCnDlUn6H6LAc8UdVJD2+HBF/Hfiq7UtLk8cXgE9KWq+st5Gkl5Tne0h6SimOdwEPlsd5wA3AIZIeJ2k1STuVXR0LvFvSZqXdvNduP+jbwVg3AnN7J5BtXw/8BPj3so9taI7yv1byvUbS7PJ79I7Ex16Ng+1FwFU0TVRjvQm4sDQj3QqsLmkrmnMU1/SttzNwuu0/TOL3iI5K4Y9afF/S3TRH6f8IfAJ4fd/y99M055wj6S6aq1x6JzC3KNP3AD8FPmv7zHKp4540J4N/TXPyde/ymqNomkvOAn4F3EdzyeRk/E/5eaukC8rzfWlOIC8GjgMOsn1aWbYbzYnZe2hO9O5j+75xtv15Hj43ADRXLQHvAj4EUD6cDgBOBw4fk3u/Mi9iXMpALBH1KJdhXgjsWi7jfCyvfSZwhO0dl0m4GBkp/BERHZOmnoiIjknhj4jomBT+iIiOSeGPiOiYKm/gmjVrlufOndt2jIiI5cbChQtvsT176WtWWvjnzp3LggUL2o4REbHckHTd0tdqpKknIqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjqmyhu4IqLx8b3HG573sXnvN06Ylu3EaMgRf0REx6TwR0R0TAp/RETHpPBHRHRMCn9ERMek8EdEdEwKf0RExyz1On5JRwF7ADfZfkaZ9w1gy7LKE4A7bG874LXXAncDDwJLbM+bptwRETFFk7mB62jgMODLvRm29+49l/Rx4M4JXr+L7VumGjAiIqbXUgu/7bMkzR20TJKAvwJeOL2xIiJiWRm2y4bnAzfavmqc5QZOlWTg87aPGHJ/EcvMZ956+rRs5+2H5zgo6jZs4d8XOHaC5TvZXixpPeA0ST+3fdagFSXNB+YDzJkzZ8hYERExnilf1SNpJeCVwDfGW8f24vLzJuA4YIcJ1j3C9jzb82bPnj3VWBERsRTDXM75Z8DPbS8atFDS4ySt1XsOvBi4bIj9RUTENFhq4Zd0LPBTYEtJiyS9sSzahzHNPJI2lHRimVwfOFvSxcB5wA9snzx90SMiYiomc1XPvuPMf92AeYuB3cvza4BnDZkvIiKmWe7cjYjomBT+iIiOSeGPiOiYjLkbM+6Kpz192rb19J9fMW3biuiKHPFHRHRMCn9ERMek8EdEdEwKf0REx6TwR0R0TAp/RETHpPBHRHRMCn9ERMek8EdEdEwKf0REx6TwR0R0TAp/RETHpPBHRHRMCn9ERMek8EdEdMxkBls/StJNki7rm3ewpN9Iuqg8dh/ntbtJulLS1ZIOnM7gERExNZM54j8a2G3A/E/a3rY8Thy7UNKKwGeAlwJbAftK2mqYsBERMbylFn7bZwG3TWHbOwBX277G9v3A14GXT2E7ERExjYZp4z9A0iWlKeiJA5ZvBFzfN72ozIuIiBZNtfB/Dtgc2Ba4Afj4gHU0YJ7H26Ck+ZIWSFpw8803TzFWREQszZQKv+0bbT9o+yHgCzTNOmMtAjbpm94YWDzBNo+wPc/2vNmzZ08lVkRETMKUCr+kDfom/wK4bMBq5wNbSNpM0irAPsDxU9lfRERMn5WWtoKkY4GdgVmSFgEHATtL2pam6eZa4C1l3Q2BL9re3fYSSQcApwArAkfZvnyZ/BYRETFpSy38tvcdMPvIcdZdDOzeN30i8KhLPSMioj25czciomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomOWWvglHSXpJkmX9c37mKSfS7pE0nGSnjDOa6+VdKmkiyQtmM7gERExNZM54j8a2G3MvNOAZ9jeBvgF8IEJXr+L7W1tz5taxIiImE5LLfy2zwJuGzPvVNtLyuQ5wMbLIFtERCwD09HG/wbgpHGWGThV0kJJ86dhXxERMaSVhnmxpH8ElgBfG2eVnWwvlrQecJqkn5dvEIO2NR+YDzBnzpxhYkVExASmfMQvaX9gD2A/2x60ju3F5edNwHHADuNtz/YRtufZnjd79uypxoqIiKWYUuGXtBvwfuBltn8/zjqPk7RW7znwYuCyQetGRMTMmczlnMcCPwW2lLRI0huBw4C1aJpvLpJ0eFl3Q0knlpeuD5wt6WLgPOAHtk9eJr9FRERM2lLb+G3vO2D2keOsuxjYvTy/BnjWUOkiImLa5c7diIiOSeGPiOiYFP6IiI5J4Y+I6JgU/oiIjknhj4jomBT+iIiOSeGPiOiYFP6IiI5J4Y+I6JgU/oiIjknhj4jomBT+iIiOSeGPiOiYFP6IiI5J4Y+I6JgU/oiIjknhj4jomBT+iIiOmVThl3SUpJskXdY3bx1Jp0m6qvx84jiv3b+sc5Wk/acreERETM1kj/iPBnYbM+9A4Ie2twB+WKYfQdI6wEHAc4AdgIPG+4CIiIiZManCb/ss4LYxs18OHFOeHwO8YsBLXwKcZvs227cDp/HoD5CIiJhBw7Txr2/7BoDyc70B62wEXN83vajMi4iIlizrk7saMM8DV5TmS1ogacHNN9+8jGNFRHTXMIX/RkkbAJSfNw1YZxGwSd/0xsDiQRuzfYTtebbnzZ49e4hYERExkWEK//FA7yqd/YHvDVjnFODFkp5YTuq+uMyLiIiWTPZyzmOBnwJbSlok6Y3AIcCLJF0FvKhMI2mepC8C2L4N+Gfg/PL4SJkXEREtWWkyK9ned5xFuw5YdwHwpr7po4CjppQuIiKmXe7cjYjomEkd8UdE9Cw68MfTtq2ND3n+tG0rJi9H/BERHZPCHxHRMSn8EREdk8IfEdExKfwRER2Twh8R0TEp/BERHZPCHxHRMSn8EREdkzt3I2K5d/DBB1e5rVrliD8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomOmXPglbSnpor7HXZL+dsw6O0u6s2+dDw8fOSIihjHlG7hsXwlsCyBpReA3wHEDVv2x7T2mup+IiJhe09XUsyvwS9vXTdP2IiJiGZmuwr8PcOw4y3aUdLGkkyRtPU37i4iIKRq6rx5JqwAvAz4wYPEFwKa275G0O/BdYItxtjMfmA8wZ86cYWNFRLTqh6dvPm3b2vWFv5y2bcH0HPG/FLjA9o1jF9i+y/Y95fmJwMqSZg3aiO0jbM+zPW/27NnTECsiIgaZjsK/L+M080h6kiSV5zuU/d06DfuMiIgpGqqpR9IawIuAt/TNeyuA7cOBvYC3SVoC3AvsY9vD7DMiIoYzVOG3/Xtg3THzDu97fhhw2DD7iIiI6ZU7dyMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOmbowi/pWkmXSrpI0oIByyXp05KulnSJpO2G3WdEREzdUIOt99nF9i3jLHspsEV5PAf4XPkZEREtmImmnpcDX3bjHOAJkjaYgf1GRMQA01H4DZwqaaGk+QOWbwRc3ze9qMyLiIgWTEdTz062F0taDzhN0s9tn9W3XANe47EzyofGfIA5c+YsdadzD/zBFOM+0rWH/Pm0bAeAg9eexm3dOX3biojoM3Tht724/LxJ0nHADkB/4V8EbNI3vTGweMB2jgCOAJg3b96jPhhiap55zDOnbVuX7n/ptG0rItozVFOPpMdJWqv3HHgxcNmY1Y4H/rpc3fNc4E7bNwyz34iImLphj/jXB46T1NvWf9s+WdJbAWwfDpwI7A5cDfweeP2Q+4yIiCEMVfhtXwM8a8D8w/ueG3j7MPuJiIjpkzt3IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6JoU/IqJjUvgjIjomhT8iomNS+CMiOiaFPyKiY1L4IyI6ZsqFX9Imks6QdIWkyyW9a8A6O0u6U9JF5fHh4eJGRMSwhhlsfQnwXtsXSFoLWCjpNNs/G7Pej23vMcR+IiJiGk35iN/2DbYvKM/vBq4ANpquYBERsWxMSxu/pLnAnwDnDli8o6SLJZ0kaevp2F9EREzdME09AEhaE/g28Le27xqz+AJgU9v3SNod+C6wxTjbmQ/MB5gzZ86wsSIiYhxDHfFLWpmm6H/N9nfGLrd9l+17yvMTgZUlzRq0LdtH2J5ne97s2bOHiRURERMY5qoeAUcCV9j+xDjrPKmsh6Qdyv5uneo+IyJieMM09ewEvBa4VNJFZd4/AHMAbB8O7AW8TdIS4F5gH9seYp8RETGkKRd+22cDWso6hwGHTXUfEREx/XLnbkREx6TwR0R0TAp/RETHpPBHRHRMCn9ERMek8EdEdEwKf0REx6TwR0R0TAp/RETHpPBHRHRMCn9ERMek8EdEdEwKf0REx6TwR0R0TAp/RETHpPBHRHRMCn9ERMek8EdEdEwKf0RExwxV+CXtJulKSVdLOnDA8lUlfaMsP1fS3GH2FxERw5ty4Ze0IvAZ4KXAVsC+krYas9obgdttPwX4JPAfU91fRERMj2GO+HcArrZ9je37ga8DLx+zzsuBY8rzbwG7StIQ+4yIiCENU/g3Aq7vm15U5g1cx/YS4E5g3SH2GRERQ5Ltqb1QehXwEttvKtOvBXaw/Y6+dS4v6ywq078s69w6YHvzgfllckvgyikFe6RZwC3TsJ3pVGMmqDNXMk1OMk1ejbmmK9OmtmdPZsWVhtjJImCTvumNgcXjrLNI0krA2sBtgzZm+wjgiCHyPIqkBbbnTec2h1VjJqgzVzJNTjJNXo252sg0TFPP+cAWkjaTtAqwD3D8mHWOB/Yvz/cCTvdUv2JERMS0mPIRv+0lkg4ATgFWBI6yfbmkjwALbB8PHAl8RdLVNEf6+0xH6IiImLphmnqwfSJw4ph5H+57fh/wqmH2MaRpbTqaJjVmgjpzJdPkJNPk1ZhrxjNN+eRuREQsn9JlQ0REx6TwR0R0zFBt/DWT9DjgPtsPtp2lp5ZMknYEXgM8H9gAuBe4DPgB8FXbd7aUaz1gJ2DDvkwLbD/URp5aM5VcKwDP6st1ue0bW8xT69+pulw1vHcj08Zf/pj7APsBzwb+AKwK3ExzAvoI21clk06iud/ie8AC4CZgNeCpwC7AnsAnylVZM5VpF+BAYB3gwjGZNqfp7uPjtu/qcqaSa3Pg/cCfAVfR/Fvq5fo98HngmJkqbBX/narLVdN7N0qF/0fA/9IUtMt6fzxJ69AUtFcDx9n+asczzbI94V2Ck1lnmjN9DDjU9q8HLFsJ2ANY0fa3u5yp7PtY4HPAj8feE1OObl9N0zHiMYNevwzy1Pp3qi5XTe/dKBX+lW0/MOw6o55pnAzr2B54R3VEjJ6RObk7UfGUtObS1lkWJrO/mc4kaSdJV0i6XNJzJJ0GLJB0fWn7r4qk17e476dJ2rX376dv/m5tZSr730HSs8vzrSS9R9LubWYapM33ruy/uvevlvduZI74JyLp17bntLDfbWhuztgIOAl4v+3by7LzbO/QQqbzaMZJWBP4PvAK22dL2o7mq/FOM51pIi2+d+8E3g5cAWwLvMv298qyC2xvN9OZyr4PohkDYyXgNOA5wJk07can2P7XNnIN0tZ7V/Zd3ftX03s3Mlf1SHrPeItoilwbPgscDJwDvAk4W9LLbP8SWLmlTCvbvhRA0s22zwawfYGk1dsIJOmS8RYB689klj5vBra3fU8ZOe5bkuba/lTJ1Za9aArZqsBvgY1t31XatM8FZrTwV/reQZ3vXzXv3cgUfuDfgI8BSwYsa6tJa03bJ5fn/ylpIXBy6cK6ra9a/X+LD4xZtspMBumzPvAS4PYx8wX8ZObjAM2Jv3sAbF8raWea4rEp7Rb+JeVy4N9L+mXvqhTb90pq4xLFGt87qPP9q+a9G6XCfwHwXdsLxy6Q9KYW8pRda+3edfG2z5D0l8C3aS4za8OHJK1h+/e2v9sXdHPgyy1lOoHmQ/KisQsknTnzcQD4raRte5nKkeMewFHAM1vKBHB/7/0Dtu/NlLQ20Ebhr/G9gzrfv2reu5Fp45e0JXDroMsQJa3fxs0tkl4NXGP7nDHz5wAfsv3mmc4UkyNpY5ojtN8OWLaT7f9rIRaSVrX9hwHzZwEb9Jrxuq7G96+m925kCn9EREzOyFzOGRERk5PCHxHRMSn8AYCkv5G0d7mdPSJG2MgX/hoLWo2ZaC5xex7wnbaD9Ej6X0knlasxqlBjJqgvV215emrM1UammgrPstIraPsBL2s5S091mWx/pu0MA/w1TbfRz207SJ8aM0F9uWrL01NjrhnPlKt6OkrS+jQ3vW1o+6WStgJ2tH1ky9EAkPREYBPb490ZOuNqzAT15aotT0+NudrKNHJNPZLWl3Skmn7nex0hvTGZHuVo4BSawSAAfgH8bWtpaG74kfR4Nd1WXwx8SdInkqn+XLXlqTlXDZlGrvBTYUGjzkyzbH+Tcseg7SVA26OVrV1uY38l8CXb29N0YJVMj1Zbrtry9NSYq/VMo1j4ayxoNWb6naR1KX0GSXou0MqQi31WkrQB8Fc0XQHUoMZMUF+u2vL01Jir9UyjWPhrLGg1ZnoPcDywuaT/o+mn5x3tRuIjNN+MrrZ9vqQn0wxRl0yPVluu2vL01Jir9Uwjd3JXpV954Bk0AyvPBvZq84ROjZlKrpWALWmuMrqy7ZHAImJmjFzhhzoLWi2ZJL1youW2W7uOX9JqNIPEbE0zCHUv0xuS6ZFqy1Vbnp4ac9WQaWSaeiS9sveguTZ+S5rR6/dcWrHrUiZgz/J4I3Akzb0E+wFfBF7TUqaerwBPounf/UfAxsDdrSaqMxPUl6u2PD015mo/k+2ReABfKo8f0AwK8e3yuA34TjI9KtsJNF3B9qY3qCDTheXnJeXnysDpyVR/rtry1Jyrhkwjc+eu7dcDSDoB2Mr2DWV6A6CVu1JrzNRnbi9PcSPNt5E29Zq/7pD0DJrh6ea2FweoMxPUl6u2PD015mo908gU/j41FrQaM50p6RTgWJqrjfYBzmg3EkeUOxk/SHPF0ZrAh9qNVGUmqC/XoDwfbjFPT425Ws80cid3JR0GbMEjC9rVtlu7VLHGTCXXK4Hnl8mzbB/XYpYVaK50+mZbGcaqMRPUmyuWHyNX+KGugtZTY6baSDrL9v9rO0e/GjNBfbkk/RvwUdt3lOknAu+1/cHkqi/TSBb+WDpJd1NuKANWoTnB9Dvbj28x04eAe4FvAL/rzbd9WzI9Um25JF1o+0/GzLvA9nZt5OnLUF2uGjKNXBt/pQWtuky21+qflvQKYIeW4vT0rmN+e988A09uIUtPjZmgvlwrqm8wcUmrA6u2lKVfjblazzRyhb/GglZjprFsf1fSgS1n2KzN/Q9SYyaoMtdXgR9K+hLNB9AbgGPajQTUmav1TJ1o6pF0ju2aBl5oPdOYG8hWAOYBL7C9YwtZXmj79PFuanMLdxPXmAnqzQUg6aXArjR3p59q+5S2svSrMVfbmUbuiH+cgtbqp1uNmWju3u1ZAlwLvLydKLwAOJ1HZuox7QwHWWMmqDcXtk8CTmpr/+OpMVfbmUbuiL98ferpFbQv2L6pnUR1ZoqYDpLOtv28MeexoDmSdVvnsWrMVVOmkSv8MTFJWwOb2z6+TH8SWLssPsz2BS1mq244yBoz1ZRL0pNtXzOT+5yMGnPVlGmUOmnbWtLL+qY/Kemo8mjl0q0aMwGHALf0Tb+Epi+hM2j/jsajqW+ksqOpLxPUk+t/ACT9sIV9T6TGXNVkGqU2/kOAf++bfgnNLexr0BS0VyQT0HTM9pO+6btsfxtA0ltayNNvlu1vSvoANCOVSWp7pLIaM0E9uVaQdBDwVEnvGbvQdlvj29aYq5pMI3PEzzgFzfZXgFnJ9EePuLR0zJVF681wlrFqHKmsxkxQT659gPtoDiLXGvBoS425qsk0Skf8NRa0GjMtlvQc2+f2zyyFY3FLmXrGDgc5G9ir3UhVZoJKctm+EvgPSZeUK1WqUGOumjKNzMldSWcAB45T0A6xvXMygaQdaG7zPxroncjdHtgf2Nv2eTOdqZ8qGamsX42ZoN5cUb9RKvzVFbQaM5Vc6wEH0Az9BnA58BnbN7aUp7rhIGvMBPXmiuXLyBR+qK+g1ZqpNn33OawH/CnNDUoAuwBn2p7xYSprzFRrLjXdRD93zPms1tWYq5ZMI1X4Y/mmZqSyN3vMSGVtFdlaM9WYS9JP2+juY2lqzFVDplG6qieWfzWOVFZjJqgv16mS/lKSWswwSI25Ws+UI/6Ok/Q4279b+prLniocqazGTDXmKt0QPI6mS5L7aLnLhppz1ZBpZAt/TQWtp6ZMkv4U+CKwpu05kp4FvMX237Scq7qRymrMBPXmivqNXOGvsaBVmulcmuu+j3cZDUjSZbaf0VamWH6N0wXJncB1tpfMdJ6eGnNVkcn2SD2Ac4FNgAv75l2WTI/OVH72Z7q45UzPBc4H7gHuBx6kuds5mSrPBZxTciwsj/tLvmuAFydXXZlG8uSu7evHzGq9b5UKM11fvolY0iqS3gdc0XKmw4B9gauA1YE3AYe2mqjOTFBfrmuBP7G9ve3tgW2By4A/Az6aXHVlGsXCX2NBqzHTW2nGa90IWETzj+/tE75iBti+GljR9oO2v0RzfXoyDVBZrqfZvrw3YftnNMWt7W6Ia8zVeqZR6qun563Ap3i4oJ1K+wWtuky2bwH2azPDAL+XtApwkaSPAjfQXP2QTI9WW64rJX0O+HqZ3hv4haRVgTa7kqgxV+uZRu7kbkxM0qFMMOyj7XfOYJxHkLQpzfXoqwDvphkg5rPlyDaZKs4laXXgb4Dn0VyeeDbwWZrLFdewfU9y1ZNpZAp/jQWt0kz7l6c7AVvR9CUE8Cpgoe13t5BpNjC7fOXtn/8M4EbbNydT3blKhlVoOo0zFXUaV2OutjONUhv/Apoz5KsB29Gc9LqKpu26rROp1WWyfYztY2hu/tnF9qG2DwV2LbnacChNt8JjbUTTRNaGGjNBpbkk7Uzzb/swmqPXX0j6f23l6akxVxWZ2ricaVk+aIYQXLlvemXgjGR6VKYrgXX6pp9Ic+TRRpbLJ1jWymWvNWaqPNdCYMu+6afSfINsJU/NuWrINIondzekGQDltjK9Jg+PS9qWGjMdAlxYxgwAeAFwcEtZVp7ismWpxkxL23erudwMNAKA7V9IajNPT425Ws80ioW/poLWU10m21+SdBLwnDLrQNu/bSnOVZJ2t31i/0xJL6W5qSWZHlZrrgWSjgS+Uqb3ozmybVuNuVrPNDInd/tJehIPF7RzWyxof1RjplpIeipwAvATHv4PMA/YEdjD9i+Sqfpcq9Jcoty7UuUsmm6i728jT825asg0koU/lj/lP8OrgV5fQZcD/237vmR6pBpzSXqX7U8tbd5MqzFXDZlS+KN1kuSl/EOczDqjnmmy+2wp1wW2txsz70KXDgDbUmOuGjKNYht/TIKkzYFFtv9QLi9Rb6fwAAAMSElEQVTbBviy7TtaiHOGpG8D37P9676Mq9B8Hd6f5sqoozueqbpckval+faxmaTj+xY9Hrh1JjIMUmOumjKN3BF/ZQWt5kwX0bQNzwVOAY6nucRs9xayrAa8geYk12bAHTQdj61A073FZ2xf1PVMNeYqdxBvBvw7cGDforuBS9xe18fV5aop0ygW/moKWuWZLrC9naS/A+6zfWjbX4FLrpWBWcC9bX4w9qsxE9SXS9L6wLPL5Hm2b2ozT0+NudrONEp37vY8VD45/wL4LzddEGyQTI/yQPnquT/NVSLQ7nXgANh+wPYNNRSynhozQV25JL0KOI+m64+/As6VtFe7qerMVUOmUWzj7y9oe5Z5bRe0GjO9nqbX0H+1/StJmwFfbTlTLL8+CDy7d+Ra+hT6X+BbraaqM1frmUax8NdY0KrL5KaTr3f2Tf+K5kaziKlYYUxzxa3U0aJQY67WM41cG39MjqQtaE4ybUXTiRwAtp/cWqhYbkn6GM1FC8eWWXvTnLB8f3up6sxVQ6aRK/w1FrRKM50NHAR8kqb56fU0/x4OaitTLN8k/SVNd98CzrJ9XMuRgDpztZ1pFAt/dQWt0kwLbW8v6VLbzyzzfmz7+W1liuWPpOfaPqftHGPVmKumTG23dS0Lq9v+IU1hvc72wcALk+lR7pO0Ak2nXwdI+gtgvZYzxfLns70nkn7aZpAxasxVTaZRPLn7iIIG/Ib2C1qNmf4WWIPmBO8/03wQ7T/hKyIeTX3PVxt3rZlXY65qMo1i4a+xoFWXyfb55ek9NE1PEVOxgqQn0rQe9J7/scDZvm3cV3YvVzWZRq6NPyYm6ftMPA7wy2YwTiznJF0LPMQjj2Z73NYFDDXmqinTyBT+GgtapZleMNFy2z+aqSwR0Y5RKvzVFbQaM/UrPTo+tUxeafuBNvNExMwYmcLfr8aCVlum0kvoMcC1NF89NwH2t31Wi7EiYgaMXOGvsaBVmmkh8GqXQZ/VDOl3rO3t28oUETNjFK/q+Tjw4rEFDWizoNWYaeVeHgDbvyjd/EY8ZjWOOVFrrhoyjeINXI8qaLTfE2aNmRZIOlLSzuXxBR4evDvisfo28KCkpwBH0gw48t/tRgLqzNV6plE84l8g6UjgK2V6P9ovaDVmehvwdpp7CwScRd+dhRGP0UO2l5Q7wP+rN7BP26GoM1frmUax8NdY0KrLZPsPwCfKI2JYNY45AXXmaj3TyJ3cjYmVnkL/EbiNpuh/AXg+8EvgTX139EZMmqStaMac+KntY8uYE3vbbnWMhxpz1ZBpZAp/jQWt0kxnA18GHg+8m6Y7ie+XXP9i+zkznSkiZtYondz9EvATYDFwLnAUzUDU7wMOS6Y/WtP2Ebb/k2aQ7v+xfZ/t04BVW8oUyzlJW0j6lqSfSbqm90iuOjONUuGvsaDVmOmhvud3TbAs4rH4EvA5YAmwC823yq9M+IqZUWOu1jONUuGvsaDVmOlpki6RdGnf8970li1liuVfjWNOQJ25Ws80Slf1PE3SJTRXzWxenlOm2xrisMZMT29pvzHaahxzAurM1XqmUTq5u+lEy21fN1NZeirNJC/lTZ/MOhH9JD0buAJ4As2YE2sDH217qMEac9WQaZQKf3UFrdJMZ9LcOfg927/um78K8Dyaa4vPsH30TGWKiJk1SoX/TCoraJVmWg14A83dw5sBd9AMA7cicCrwGdsXzVSeGA2l/6m/AzalrwnZdqvt6TXmqiHTKBX+6gpajZnG5FuZ5vLSe9vuTCuWb5IuBg6n6Yrkwd582612TVJjrhoyjUzh71djQasxU8R0kbSwxi69a8xVQ6aRLPwRMbMkHQzcBBwH/KE33+0Ntg7UmauGTCn8ETE0Sb8aMNtuabD1nhpz1ZAphT8iomNG6QauiJhhkl5o+3RJrxy03PZ3ZjoT1Jmrpkwp/BExjBcAp/Nwv/L9DLRS+KkzVzWZ0tQTEdExOeKPiKFJ+vCg+bY/MtNZ+tWYq4ZMKfwRMR1+1/d8NWAPmv5o2lZjrtYzpaknIqadpFWB422/pO0s/WrM1UamUeqPPyLqsQbtdT0+kRpzzXimNPVExNDKQD695oMVgdlAq+37UGeuGjKlqScihjZm7IklwI22l7SVp6fGXDVkSuGPiKFJWmei5W31jVNjrhoypfBHxNAkXQtsAtxOM7ToE4DeGBSt9Y1TY64aMuXkbkRMh5OBPW3Psr0uzSWK37G9WcsdtdWYq/VMOeKPiKEN6mNe0gLb89rKVDJUl6uGTLmqJyKmwy2SPgh8leaKldcAt7YbCagzV+uZ0tQTEdNhX5rLEo8DvgusV+a1rcZcrWdKU09ERMekqScihiZpNvD3wNY0/c8AYPuFrYWizlw1ZEpTT0RMh68BPwc2A/4JuBY4v81ARY25Ws+Upp6IGFrvShVJl9jepsz7ke0XJFd9mdLUExHT4YHy8wZJfw4sBjZuMU9Pjblaz5TCHxHT4V8krQ28FzgUeDzw7nYjAXXmaj1TmnoiIjomR/wRMWWSPj3RctvvnKks/WrMVVOmFP6IGMZbgcuAb9K0VavdOH9UY65qMqWpJyKmTNK6wKuAvWn6lv8G8G3btydXvZlyHX9ETJntW20fbnsX4HU0XQxfLum1yVVvpjT1RMTQJG1H09/Mi4CTgIXtJmrUmKuGTGnqiYgpk/RPNP3JXwF8HTi57aENoc5cNWVK4Y+IKZP0EHANcG+Z1SsoohlNapvkqi9TmnoiYhibtR1gHDXmqiZTjvgjYsokyUspIpNZZ7rVmKumTLmqJyKGcYakd0ia0z9T0iqSXijpGGD/5KorU474I2LKJK0GvAHYj6Yp4w6aPuZXBE4FPmP7ouSqK1MKf0RMC0krA7OAe23f0XaenhpztZ0phT8iomPSxh8R0TEp/BERHZPCHzNK0oOSLpJ0uaSLJb1H0jL/dyjpdZI2XNb7mWD/r5C01WN8zcGS3leet5o/RksKf8y0e21va3trmr5KdgcOWpY7lLQiTadYbRbOVwCPqfCP8TrazR8jJIU/WmP7JmA+cIAaK0r6mKTzJV0i6S0AknaWdJak4yT9TNLhvW8Jkj4naUH5BvFPvW1LulbShyWdTdMh1jzga+Xbxupl2fmSLpN0hCSV150p6ZNlf1dIerak70i6StK/9G3/NZLOK9v7fPlwQdI9kv61fJs5R9L6kv4UeBnwsbL+5pK2LcsvKb/XE8f7O0naa0D+7SX9SNJCSadI2uCx5I+Os51HHjP2AO4ZMO92YH2aD4EPlnmrAgtornfeGbgPeDLNNc+nAXuV9dYpP1cEzgS2KdPXAn/ft48zgXl90+v0Pf8KsGffev9Rnr+LZsCMDUqeRcC6wNOB7wMrl/U+C/x1ee6+bX207/c5upe5TF8CvKA8/wjwXwP+LgcD7xubH1gZ+Akwu0zvDRw12fxt/xvIo/1H+uqJGvRGInoxsE05wgVYG9gCuB84z/Y1AJKOBZ4HfAv4K0nzafqd2oCmOeWS8vpvTLDPXST9PbAGsA5wOU0xBzi+/LwUuNz2DWW/1wCblH1vD5xfviisDtxUXnM/cEJ5vpCmOeuRv2wz0PYTbP+ozDoG+J8Jso61JfAM4LSy/xWBG/qWLy3/rY9hXzGCUvijVZKeDDxIUzgFvMP2KWPW2ZmHezLssaTNgPcBz7Z9u6Sjae6E7PndOPtcjeYofZ7t6yUdPOZ1fyg/H+p73pteqeQ8xvYHBmz+Adu9rA+ybP6Piaag7zjO8qXlj45LG3+0RtJs4HDgsFIsTwHeVu5qRNJTJT2urL6DpM1K2/7ewNnA42mK+52S1gdeOsHu7gbWKs97Rf4WSWsCew1+ybh+COwlab2Scx1Jmy7lNX/cv+07gdslPb8sey3wo7KtAyQdsJT8VwKzJe1YXrOypK0f4+8QHZZP/5hpq0u6iKadeglN+/onyrIvAnOBC8rJ1ptproYB+ClwCPBM4CzgONsPSbqQppnmGuD/Jtjv0cDhku4FdgS+QNMUci1w/mP5BWz/TNIHgVPLB9EDwNuB6yZ42deBL0h6J80Hzf4lzxol++vLek8b5/cYm38v4NOl2Wgl4L9o/g4RS5UuG6J6pannfbb3aDvLsibpBOCVtu9vO0uMrhzxR1SkCx9u0b4c8UdEdExO7kZEdEwKf0REx6TwR0R0TAp/RETHpPBHRHRMCn9ERMf8f1H31ccFthn6AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp['DescProd_Vendas'] = temp.Desconto_Produto/temp.Vendas\n",
    "temp['DescFret_Vendas'] = temp.Desconto_Frete/temp.Vendas\n",
    "temp['DescTot_Vendas'] = temp.Desconto_Total/temp.Vendas\n",
    "temp['Desconto_no_preco'] = temp.DescTot_Vendas/temp.Preco_Medio * 100\n",
    "\n",
    "temp.drop(['Desconto_Produto','Desconto_Frete','Desconto_Total','Vendas','Preco_Medio'], axis=1, inplace=True)\n",
    "temp.set_index(['Departamento','Item']).plot(y='Desconto_no_preco', kind='bar', legend=None, title='Descontos (%)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "De acordo com os dados observados, os descontos no departamento de bolas de gude giram em torno de 5%, com praticamente nenhuma diferenciação entre os itens. O departamento de cadeiras tem uma política de descontos mais agressiva, com descontos de 17% em média. Neste departamento existe uma diferenciação maior entre os itens, sendo que o item 3 foi o que acumulou mais descontos (18,5%). No departamento de máquinas fotográficas, os descontos médios estão na faixa de 8,5%, sendo o item 1 o que recebeu mais descontos totais (aprox. 9,5%). Precisamos ver o volume de vendas de cada item para saber se os descontos estão sendo bem aplicados."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Agora, vamos ver as métricas de vendas e visitas. Vamos ver como foi o tráfego no período analisado."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Visitas</th>\n",
       "      <th>Vendas</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mes</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2018-06-01</th>\n",
       "      <td>181075</td>\n",
       "      <td>8648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-07-01</th>\n",
       "      <td>181196</td>\n",
       "      <td>19520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-08-01</th>\n",
       "      <td>176556</td>\n",
       "      <td>15991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2018-09-01</th>\n",
       "      <td>180264</td>\n",
       "      <td>10484</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Visitas  Vendas\n",
       "Mes                        \n",
       "2018-06-01   181075    8648\n",
       "2018-07-01   181196   19520\n",
       "2018-08-01   176556   15991\n",
       "2018-09-01   180264   10484"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby('Mes')['Visitas','Vendas'].sum()\n",
    "temp\n",
    "#temp.plot(kind='line', title='Visitas e vendas')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "De acordo com a tabela, ao longo dos meses de junho a setembro as visitas se mantém estáveis em torno de 180.000, enquanto as vendas começam em 8648 para mais que dobrar no mês de julho, para depois tomar uma tendência de queda nos meses seguintes. Dados de frequência diária poderiam revelar a influência de datas comemorativas no tráfego do site.\n",
    "\n",
    "Podemos calcular a taxa de conversão dos itens através da relação:\n",
    "\n",
    "$\\textrm{Taxa de conversão} = \\frac{\\textrm{Vendas}}{\\textrm{Visitas}} * 100$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f07598be48>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp = df.groupby(['Departamento','Item'])['Visitas','Vendas','Custo_Mkt','Preco_Medio'].mean().round(2)\n",
    "temp.reset_index()\n",
    "temp['Taxa de conversão'] = temp.Vendas / temp.Visitas * 100\n",
    "\n",
    "temp['Taxa de conversão'].plot(kind='bar', title='Taxa de conversão (%)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Descobrimos que existe uma grande variabilidade entre os itens. No departamento de bolas de gude, o item 1 é o que possui a maior taxa de conversão (requer o menor número de visitas por venda) bem como o item 1 do setor de Cadeiras e o item 1 do setor de máquinas fotográficas. Estratégias de marketing que busquem aumentar a frequência dos consumidores na web podem focar nesses itens, pois requerem o menor investimento para alavancar vendas.\n",
    "\n",
    "Sabe-se que os dispêndios com marketing objetivam ampliar as vendas dos itens oferecidos. Na tabela a seguir, calculamos a participação do custo médio de marketing no preço médio dos itens, calculando o gasto médio por item em marketing (representado pela coluna `Custo_Mkt_Vendas`), e em seguida a participação desse custo no preço médio dos itens, através das relações:\n",
    "\n",
    "$\\textrm{Custo_Mkt_Vendas} = \\frac{C_{mkt}}{\\textrm{Vendas}}$\n",
    "\n",
    "$\\textrm{Custo_Mkt_Preço} = \\frac{\\textrm{Custo_Mkt_Vendas}}{P_{médio}}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Taxa de conversão</th>\n",
       "      <th>Custo_Mkt_Vendas</th>\n",
       "      <th>Custo_Mkt_Preço</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>10.52</td>\n",
       "      <td>1.72</td>\n",
       "      <td>10.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9.80</td>\n",
       "      <td>1.70</td>\n",
       "      <td>10.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>8.58</td>\n",
       "      <td>1.72</td>\n",
       "      <td>10.37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>6.59</td>\n",
       "      <td>40.71</td>\n",
       "      <td>14.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>5.66</td>\n",
       "      <td>41.07</td>\n",
       "      <td>11.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.55</td>\n",
       "      <td>43.15</td>\n",
       "      <td>9.80</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>8.69</td>\n",
       "      <td>89.38</td>\n",
       "      <td>6.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8.03</td>\n",
       "      <td>90.80</td>\n",
       "      <td>4.59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6.79</td>\n",
       "      <td>88.14</td>\n",
       "      <td>2.33</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                            Taxa de conversão  Custo_Mkt_Vendas  \\\n",
       "Departamento          Item                                        \n",
       "Bolas de Gude         1                 10.52              1.72   \n",
       "                      2                  9.80              1.70   \n",
       "                      3                  8.58              1.72   \n",
       "Cadeiras              1                  6.59             40.71   \n",
       "                      2                  5.66             41.07   \n",
       "                      3                  4.55             43.15   \n",
       "Maquinas Fotogr ficas 1                  8.69             89.38   \n",
       "                      2                  8.03             90.80   \n",
       "                      3                  6.79             88.14   \n",
       "\n",
       "                            Custo_Mkt_Preço  \n",
       "Departamento          Item                   \n",
       "Bolas de Gude         1               10.41  \n",
       "                      2               10.75  \n",
       "                      3               10.37  \n",
       "Cadeiras              1               14.47  \n",
       "                      2               11.47  \n",
       "                      3                9.80  \n",
       "Maquinas Fotogr ficas 1                6.61  \n",
       "                      2                4.59  \n",
       "                      3                2.33  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp['Custo_Mkt_Vendas'] = temp.Custo_Mkt / temp.Vendas\n",
    "temp['Custo_Mkt_Preço'] = temp.Custo_Mkt_Vendas / temp.Preco_Medio * 100\n",
    "temp.drop(['Visitas','Vendas','Custo_Mkt','Preco_Medio'], axis=1).round(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1f0759fa4a8>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "temp.unstack()['Custo_Mkt_Preço'].plot(kind='bar', title='Custos de marketing em relação ao preço médio (%)')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A partir dos dados calculados, vemos que no departamento de bolas de gude os gastos com marketing chegam ao redor de 10,5% em média. No departamento de cadeiras, esse número alcança aproximadamente 12%. O item 1 recebe o maior gasto em marketing (14,4%). O departamento de máquinas fotográficas é o que possui os menores gastos médios (4,5%)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Análise regional\n",
    "\n",
    "Nesta seção avaliaremos diferenças regionais nas principais métricas de desempenho da empresa. Vamos mostrar primeiro os preços, custos marginais e margens de lucro médias:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Preco_Medio</th>\n",
       "      <th>Custo_Marginal</th>\n",
       "      <th>Markup</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UF</th>\n",
       "      <th>Departamento</th>\n",
       "      <th>Item</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"9\" valign=\"top\">MT</th>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>17.75</td>\n",
       "      <td>15.26</td>\n",
       "      <td>14.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>15.12</td>\n",
       "      <td>13.03</td>\n",
       "      <td>14.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>17.89</td>\n",
       "      <td>15.45</td>\n",
       "      <td>14.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>271.16</td>\n",
       "      <td>220.83</td>\n",
       "      <td>18.58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>345.75</td>\n",
       "      <td>275.56</td>\n",
       "      <td>19.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>421.02</td>\n",
       "      <td>392.71</td>\n",
       "      <td>5.86</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>1400.65</td>\n",
       "      <td>1361.26</td>\n",
       "      <td>2.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1828.76</td>\n",
       "      <td>1909.82</td>\n",
       "      <td>-4.67</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3360.61</td>\n",
       "      <td>2956.03</td>\n",
       "      <td>12.49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"9\" valign=\"top\">PE</th>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>18.05</td>\n",
       "      <td>11.70</td>\n",
       "      <td>35.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>18.15</td>\n",
       "      <td>11.78</td>\n",
       "      <td>35.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>18.11</td>\n",
       "      <td>11.75</td>\n",
       "      <td>35.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>282.86</td>\n",
       "      <td>242.73</td>\n",
       "      <td>13.52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>355.98</td>\n",
       "      <td>298.36</td>\n",
       "      <td>15.34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>436.64</td>\n",
       "      <td>421.18</td>\n",
       "      <td>2.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>1208.30</td>\n",
       "      <td>1200.00</td>\n",
       "      <td>-0.76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2098.28</td>\n",
       "      <td>1733.88</td>\n",
       "      <td>17.45</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4101.21</td>\n",
       "      <td>3663.38</td>\n",
       "      <td>10.64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"9\" valign=\"top\">SP</th>\n",
       "      <th rowspan=\"3\" valign=\"top\">Bolas de Gude</th>\n",
       "      <th>1</th>\n",
       "      <td>13.70</td>\n",
       "      <td>9.30</td>\n",
       "      <td>31.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>14.12</td>\n",
       "      <td>9.61</td>\n",
       "      <td>31.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>13.83</td>\n",
       "      <td>9.37</td>\n",
       "      <td>31.79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Cadeiras</th>\n",
       "      <th>1</th>\n",
       "      <td>289.90</td>\n",
       "      <td>216.53</td>\n",
       "      <td>24.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>372.32</td>\n",
       "      <td>274.29</td>\n",
       "      <td>25.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>463.33</td>\n",
       "      <td>397.47</td>\n",
       "      <td>13.32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">Maquinas Fotogr ficas</th>\n",
       "      <th>1</th>\n",
       "      <td>1450.30</td>\n",
       "      <td>1339.26</td>\n",
       "      <td>7.22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2006.28</td>\n",
       "      <td>1711.73</td>\n",
       "      <td>14.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3907.90</td>\n",
       "      <td>3219.24</td>\n",
       "      <td>17.47</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                               Preco_Medio  Custo_Marginal  Markup\n",
       "UF Departamento          Item                                     \n",
       "MT Bolas de Gude         1           17.75           15.26   14.03\n",
       "                         2           15.12           13.03   14.03\n",
       "                         3           17.89           15.45   14.03\n",
       "   Cadeiras              1          271.16          220.83   18.58\n",
       "                         2          345.75          275.56   19.30\n",
       "                         3          421.02          392.71    5.86\n",
       "   Maquinas Fotogr ficas 1         1400.65         1361.26    2.40\n",
       "                         2         1828.76         1909.82   -4.67\n",
       "                         3         3360.61         2956.03   12.49\n",
       "PE Bolas de Gude         1           18.05           11.70   35.07\n",
       "                         2           18.15           11.78   35.07\n",
       "                         3           18.11           11.75   35.07\n",
       "   Cadeiras              1          282.86          242.73   13.52\n",
       "                         2          355.98          298.36   15.34\n",
       "                         3          436.64          421.18    2.64\n",
       "   Maquinas Fotogr ficas 1         1208.30         1200.00   -0.76\n",
       "                         2         2098.28         1733.88   17.45\n",
       "                         3         4101.21         3663.38   10.64\n",
       "SP Bolas de Gude         1           13.70            9.30   31.79\n",
       "                         2           14.12            9.61   31.79\n",
       "                         3           13.83            9.37   31.79\n",
       "   Cadeiras              1          289.90          216.53   24.50\n",
       "                         2          372.32          274.29   25.70\n",
       "                         3          463.33          397.47   13.32\n",
       "   Maquinas Fotogr ficas 1         1450.30         1339.26    7.22\n",
       "                         2         2006.28         1711.73   14.13\n",
       "                         3         3907.90         3219.24   17.47"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['UF','Departamento','Item'])['Preco_Medio','Custo_Marginal','Markup',].mean().round(2)\n",
    "temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Percebe-se que alguns departamentos desempenham melhor em determinadas regiões. No Mato Grosso, as margens médias estão abaixo da média da empresa, sobretudo no departamento de máquinas fotográficas, que opera aproximadamente à preço de custo. O estado de Pernambuco apresenta ótimas margens no departamento de bolas de gude, enquanto que no departamento de cadeiras as margens estão abaixo da média. São Paulo apresenta as melhores margens no departamento de cadeiras e máquinas fotográficas. Nota-se a necessidade, em alguns itens, de reajuste de preços de produtos ou fretes (ou ainda, redução de descontos).\n",
    "\n",
    "A seguir, o somatório das vendas e do lucro líquido no período:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th>Vendas</th>\n",
       "      <th>Lucro_Liquido</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>UF</th>\n",
       "      <th>Departamento</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">MT</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <td>6803</td>\n",
       "      <td>10241.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cadeiras</th>\n",
       "      <td>2634</td>\n",
       "      <td>-48790.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "      <td>2194</td>\n",
       "      <td>-180710.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">PE</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <td>11321</td>\n",
       "      <td>61403.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cadeiras</th>\n",
       "      <td>4976</td>\n",
       "      <td>-195161.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "      <td>2061</td>\n",
       "      <td>-10480.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">SP</th>\n",
       "      <th>Bolas de Gude</th>\n",
       "      <td>10505</td>\n",
       "      <td>39392.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Cadeiras</th>\n",
       "      <td>10081</td>\n",
       "      <td>9467.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Maquinas Fotogr ficas</th>\n",
       "      <td>4068</td>\n",
       "      <td>485970.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          Vendas  Lucro_Liquido\n",
       "UF Departamento                                \n",
       "MT Bolas de Gude            6803        10241.4\n",
       "   Cadeiras                 2634       -48790.0\n",
       "   Maquinas Fotogr ficas    2194      -180710.0\n",
       "PE Bolas de Gude           11321        61403.0\n",
       "   Cadeiras                 4976      -195161.0\n",
       "   Maquinas Fotogr ficas    2061       -10480.0\n",
       "SP Bolas de Gude           10505        39392.0\n",
       "   Cadeiras                10081         9467.0\n",
       "   Maquinas Fotogr ficas    4068       485970.0"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['UF','Departamento'])['Vendas','Lucro_Liquido',].sum().round(2)\n",
    "temp"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Aqui emergem as diferenças de desempenho dos departamentos, nas diferentes localidades. Em Pernambuco, novamente o departamento de bolas de gude se destaca, bem como um desempenho no departamento de cadeiras melhor do que no Mato Grosso. O estado de São Paulo, dotado de um grande mercado consumidor, vendeu quase o mesmo em máquinas fotográficas que nos dois outros estados, para um lucro de quase R\\\\$ 500 mil. Ademais, novamente mostra seu potencial no departamento de cadeiras.\n",
    "\n",
    "É intuitivo imaginar que a rentabilidade dos fretes sejam sensíveis à localização dos consumidores. Seguem os resultados do lucro médio dos fretes em cada região e departamento, dividido pelas vendas:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>UF</th>\n",
       "      <th>Departamento</th>\n",
       "      <th>Lucro_Frete_Ponderado</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MT</td>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>1.09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>MT</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>-6.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>MT</td>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>-72.44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>PE</td>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>3.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>PE</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>-6.89</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>PE</td>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>-150.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>SP</td>\n",
       "      <td>Bolas de Gude</td>\n",
       "      <td>2.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>SP</td>\n",
       "      <td>Cadeiras</td>\n",
       "      <td>2.50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>SP</td>\n",
       "      <td>Maquinas Fotogr ficas</td>\n",
       "      <td>-122.03</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   UF           Departamento  Lucro_Frete_Ponderado\n",
       "0  MT          Bolas de Gude                   1.09\n",
       "1  MT               Cadeiras                  -6.63\n",
       "2  MT  Maquinas Fotogr ficas                 -72.44\n",
       "3  PE          Bolas de Gude                   3.17\n",
       "4  PE               Cadeiras                  -6.89\n",
       "5  PE  Maquinas Fotogr ficas                -150.63\n",
       "6  SP          Bolas de Gude                   2.24\n",
       "7  SP               Cadeiras                   2.50\n",
       "8  SP  Maquinas Fotogr ficas                -122.03"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp = df.groupby(['UF','Departamento'])['Lucro_Frete','Vendas'].mean().reset_index().round(2)\n",
    "temp['Lucro_Frete_Ponderado'] = temp.Lucro_Frete / temp.Vendas\n",
    "temp.drop(['Lucro_Frete','Vendas'], axis=1, inplace=True)\n",
    "temp.round(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se interpretar essa relação como um índice. Analisando os dados da tabela, nota-se que cada estado tem vantagens comparativas na parte logística, de acordo com cada departamento. Uma análise prévia não encontrou diferenças significativas entre itens de um mesmo departamento. No Mato Grosso, as perdas com frete são menores no departamento de máquinas fotográficas. Em Pernambuco, as vantagens aparecem no departamento de bolas de gude. O estado de São Paulo apresenta bons números em geral, e mais uma vez mostrando vantagens comparativas no departamento de cadeiras. É provável que aqui se encontre algum centro de distribuição destes itens.\n",
    "\n",
    "Os números apresentados nesta seção revelam a importância da logística no funcionamento da empresa, sobretudo para os itens de aparentemente difícil transporte como os do departamento de cadeiras e máquinas fotográficas."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}