LPG_Modeldrivenmethods.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "LPG_Modeldrivenmethods.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "authorship_tag": "ABX9TyOQmaTWnk8VJIztPg0zCBmS",
      "include_colab_link": true
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github",
        "colab_type": "text"
      },
      "source": [
        "<a href=\"https://colab.research.google.com/github/Anusha-T8/LPG_SALES_FORECAST/blob/main/LPG_Modeldrivenmethods.ipynb\" target=\"_parent\"><img src=\"https://colab.research.google.com/assets/colab-badge.svg\" alt=\"Open In Colab\"/></a>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Osd3-Lfp_B-y"
      },
      "source": [
        "import pandas as pd\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "%matplotlib inline\n",
        " \n",
        "import seaborn as sns"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "nkzk8bZS_Sk_"
      },
      "source": [
        "df=pd.read_csv(\"/content/total-liquefied-petroleum-gas-sales-quarterly.csv\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "evuVSvzB_jPD"
      },
      "source": [
        "del df['level_1']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "UPo6LkQWAQPw",
        "outputId": "6f6eeaa8-39ac-472e-af2a-aba7f3423e63"
      },
      "source": [
        "df.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>quarter</th>\n",
              "      <th>value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1994-Q1</td>\n",
              "      <td>29226.3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1994-Q2</td>\n",
              "      <td>38153.9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1994-Q3</td>\n",
              "      <td>39414.7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1994-Q4</td>\n",
              "      <td>37600.3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1995-Q1</td>\n",
              "      <td>34100.6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   quarter    value\n",
              "0  1994-Q1  29226.3\n",
              "1  1994-Q2  38153.9\n",
              "2  1994-Q3  39414.7\n",
              "3  1994-Q4  37600.3\n",
              "4  1995-Q1  34100.6"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "aD7JfQCuApTY",
        "outputId": "43df3a2c-cf92-46db-f09a-e9e5d4a9d4fa"
      },
      "source": [
        "df.shape"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(102, 3)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qXNK1-AkASY-"
      },
      "source": [
        "Quarter=['Q1','Q2','Q3','Q4']\n",
        "n=df['quarter'][0]\n",
        "n[5:]\n",
        "df['Quarter']=0"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jypRIBYQAbz3",
        "outputId": "1b297772-4d03-4861-87d4-269626091d07"
      },
      "source": [
        "for i in range(102):\n",
        "    n=df['quarter'][i]\n",
        "    df['Quarter'][i]=n[5:]"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/ipykernel_launcher.py:3: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  This is separate from the ipykernel package so we can avoid doing imports until\n",
            "/usr/local/lib/python3.6/dist-packages/pandas/core/indexing.py:670: SettingWithCopyWarning: \n",
            "A value is trying to be set on a copy of a slice from a DataFrame\n",
            "\n",
            "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
            "  iloc._setitem_with_indexer(indexer, value)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Mz8HV8ZaAzrd"
      },
      "source": [
        "dummy=pd.DataFrame(pd.get_dummies(df['Quarter']))\n",
        "df1=pd.concat((df,dummy),axis=1)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "1HKtqYDUBElh",
        "outputId": "cf5eb32b-1f14-4081-9939-89aefdf011c4"
      },
      "source": [
        "df1.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>quarter</th>\n",
              "      <th>value</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Q1</th>\n",
              "      <th>Q2</th>\n",
              "      <th>Q3</th>\n",
              "      <th>Q4</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1994-Q1</td>\n",
              "      <td>29226.3</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1994-Q2</td>\n",
              "      <td>38153.9</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1994-Q3</td>\n",
              "      <td>39414.7</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1994-Q4</td>\n",
              "      <td>37600.3</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1995-Q1</td>\n",
              "      <td>34100.6</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   quarter    value Quarter  Q1  Q2  Q3  Q4\n",
              "0  1994-Q1  29226.3      Q1   1   0   0   0\n",
              "1  1994-Q2  38153.9      Q2   0   1   0   0\n",
              "2  1994-Q3  39414.7      Q3   0   0   1   0\n",
              "3  1994-Q4  37600.3      Q4   0   0   0   1\n",
              "4  1995-Q1  34100.6      Q1   1   0   0   0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 19
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "z4B_h3KHBHDC"
      },
      "source": [
        "t= np.arange(1,103)\n",
        "df1['t']=t\n",
        "df1['t_square']=df1['t']*df1['t']"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FPjBqUbwB270"
      },
      "source": [
        "log_value=np.log(df1['value'])\n",
        "df1['log_value']=log_value"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "vkTBV-6ECHTq",
        "outputId": "ecccbf18-4cc8-4f59-f25a-e87a1ee4c815"
      },
      "source": [
        "df1.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>quarter</th>\n",
              "      <th>value</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Q1</th>\n",
              "      <th>Q2</th>\n",
              "      <th>Q3</th>\n",
              "      <th>Q4</th>\n",
              "      <th>t</th>\n",
              "      <th>t_square</th>\n",
              "      <th>log_value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>1994-Q1</td>\n",
              "      <td>29226.3</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>10.282824</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>1994-Q2</td>\n",
              "      <td>38153.9</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>10.549383</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>1994-Q3</td>\n",
              "      <td>39414.7</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>9</td>\n",
              "      <td>10.581894</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>1994-Q4</td>\n",
              "      <td>37600.3</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>16</td>\n",
              "      <td>10.534767</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>1995-Q1</td>\n",
              "      <td>34100.6</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>5</td>\n",
              "      <td>25</td>\n",
              "      <td>10.437070</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   quarter    value Quarter  Q1  Q2  Q3  Q4  t  t_square  log_value\n",
              "0  1994-Q1  29226.3      Q1   1   0   0   0  1         1  10.282824\n",
              "1  1994-Q2  38153.9      Q2   0   1   0   0  2         4  10.549383\n",
              "2  1994-Q3  39414.7      Q3   0   0   1   0  3         9  10.581894\n",
              "3  1994-Q4  37600.3      Q4   0   0   0   1  4        16  10.534767\n",
              "4  1995-Q1  34100.6      Q1   1   0   0   0  5        25  10.437070"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "T1Bqg20UCKcA"
      },
      "source": [
        "df1.to_excel(\"requiredforprediction.xlsx\")"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 282
        },
        "id": "yOe_eadMCkIq",
        "outputId": "16bbb232-a7ac-489a-da60-9a5eff7f6f77"
      },
      "source": [
        "df1.value.plot()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7f158e662668>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 30
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "DNN43iyeCz3U"
      },
      "source": [
        "Train=df1.head(81)\n",
        "Test=df1.tail(21)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WarXpMCRDL6a",
        "outputId": "0fce196a-bbee-4f72-b3e4-10a2c4fd2603"
      },
      "source": [
        "import statsmodels.formula.api as smf"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.6/dist-packages/statsmodels/tools/_testing.py:19: FutureWarning: pandas.util.testing is deprecated. Use the functions in the public API at pandas.testing instead.\n",
            "  import pandas.util.testing as tm\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "C7VtUaSaLeHk"
      },
      "source": [
        "LINEAR MODEL"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "CrNnCAgnDSf3",
        "outputId": "c2bfc535-61c2-4db9-e487-cb6d0fa30734"
      },
      "source": [
        "linear= smf.ols('value~t',data=Train).fit()\n",
        "predlin=pd.Series(linear.predict(pd.DataFrame(Test['t'])))\n",
        "rmselin=np.sqrt((np.mean(np.array(Test['value'])-np.array(predlin))**2))\n",
        "rmselin"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "979.9593361509145"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 34
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "bb1sK2szLhqc"
      },
      "source": [
        "QUADRATIC MODEL"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2MgBigenDolC",
        "outputId": "b9a2a937-481b-4d60-f213-2c092718b7c2"
      },
      "source": [
        "quad=smf.ols('value~t+t_square',data=Train).fit()\n",
        "predquad=pd.Series(quad.predict(pd.DataFrame(Test[['t','t_square']])))\n",
        "rmsequad=np.sqrt(np.mean((np.array(Test['value'])-np.array(predquad))**2))\n",
        "rmsequad"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2370.9543570670867"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 35
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Kr8M5p5nLk6A"
      },
      "source": [
        "EXPONENTIAL MODEL"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ezTxE6I7Dy1G",
        "outputId": "c385fa92-712a-410e-9fe4-5d53af35ca9e"
      },
      "source": [
        "expo=smf.ols('log_value~t',data=Train).fit()\n",
        "predexp=pd.Series(expo.predict(pd.DataFrame(Test['t'])))\n",
        "rmseexpo=np.sqrt(np.mean((np.array(Test['value'])-np.array(np.exp(predexp)))**2))\n",
        "rmseexpo"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1620.9259461530353"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 36
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Hu4ZIMH3Lo9H"
      },
      "source": [
        "ADDITIVE SEASONALITY "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "1Ep2FCYhD7wT",
        "outputId": "7f272c1b-3b0d-437e-f61e-316a5b6e8da4"
      },
      "source": [
        "additive= smf.ols('value~ Q1+Q2+Q3+Q4',data=Train).fit()\n",
        "predadd=pd.Series(additive.predict(pd.DataFrame(Test[['Q1','Q2','Q3','Q4']])))\n",
        "rmseadd=np.sqrt(np.mean((np.array(Test['value'])-np.array(predadd))**2))\n",
        "rmseadd"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "6902.612572520001"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 37
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2e2TmlS1Ls2F"
      },
      "source": [
        "ADDITIVE SEASONALITY WITH LINEAR TREND"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ipqZ7PfMEEsp",
        "outputId": "1f96dbc2-8609-41f7-f156-326f3c0dfa81"
      },
      "source": [
        "addlinear= smf.ols('value~t+Q1+Q2+Q3+Q4',data=Train).fit()\n",
        "predaddlinear=pd.Series(addlinear.predict(pd.DataFrame(Test[['t','Q1','Q2','Q3','Q4']])))\n",
        "rmseaddlinear=np.sqrt(np.mean((np.array(Test['value'])-np.array(predaddlinear))**2))\n",
        "rmseaddlinear"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1303.766907108997"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 38
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vh8ljCeaLwbW"
      },
      "source": [
        "ADDITIVE SEASONALITY WITH QUADRATIC TREND"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "m63ahGEJEN-s",
        "outputId": "7371a3dd-9b20-4289-9483-7691d66408e3"
      },
      "source": [
        "addquad=smf.ols('value~t+t_square+Q1+Q2+Q3+Q4',data=Train).fit()\n",
        "predaddquad=pd.Series(addquad.predict(pd.DataFrame(Test[['t','t_square','Q1','Q2','Q3','Q4']])))\n",
        "rmseaddquad=np.sqrt(np.mean((np.array(Test['value'])-np.array(predaddquad))**2))\n",
        "rmseaddquad"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2598.575351944335"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 39
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dW1p1O0XL0HM"
      },
      "source": [
        "MULTIPLICATIVE SEASONALITY"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "2J1gzI2ZEfKW",
        "outputId": "f7bbea56-7d44-4038-a2ef-d91595ec2470"
      },
      "source": [
        "mullin= smf.ols('log_value~t+Q1+Q2+Q3+Q4',data=Train).fit()\n",
        "predmullin= pd.Series(mullin.predict(pd.DataFrame(Test[['t','Q1','Q2','Q3','Q4']])))\n",
        "rmsemulin=np.sqrt(np.mean((np.array(Test['value'])-np.array(np.exp(predmullin)))**2))\n",
        "rmsemulin"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1652.464094866996"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 42
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "sR0-ptq8L4vG"
      },
      "source": [
        "MULTIPLICATIVE SEASONALITY WITH QUADRATIC TREND"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "IR02BWaLEvak",
        "outputId": "435a6423-583c-414a-d6f4-00449b9dc433"
      },
      "source": [
        "mul_quad= smf.ols('log_value~t+t_square+Q1+Q2+Q3+Q4',data=Train).fit()\n",
        "pred_mul_quad= pd.Series(mul_quad.predict(Test[['t','t_square','Q1','Q2','Q3','Q4']]))\n",
        "rmse_mul_quad=np.sqrt(np.mean((np.array(Test['value'])-np.array(np.exp(pred_mul_quad)))**2))\n",
        "rmse_mul_quad"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "2419.8544096619826"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 284
        },
        "id": "lkZc12DnE-Vp",
        "outputId": "1a3d2a2f-05d1-4101-bf63-5e409e8f45fa"
      },
      "source": [
        "Table={'Model':pd.Series(['rmselin','rmsequad','rmseexpo','rmseadd','rmseaddlinear','rmseaddquad','rmsemulin','rmse_mul_quad']),'Values':pd.Series([rmselin,rmsequad,rmseexpo,rmseadd,rmseaddlinear,rmseaddquad,rmsemulin,rmse_mul_quad])}\n",
        "RMSE_Table=pd.DataFrame(Table)\n",
        "RMSE_Table"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Model</th>\n",
              "      <th>Values</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>rmselin</td>\n",
              "      <td>979.959336</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>rmsequad</td>\n",
              "      <td>2370.954357</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>rmseexpo</td>\n",
              "      <td>1620.925946</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>rmseadd</td>\n",
              "      <td>6902.612573</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>rmseaddlinear</td>\n",
              "      <td>1303.766907</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>rmseaddquad</td>\n",
              "      <td>2598.575352</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>rmsemulin</td>\n",
              "      <td>1652.464095</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>rmse_mul_quad</td>\n",
              "      <td>2419.854410</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "           Model       Values\n",
              "0        rmselin   979.959336\n",
              "1       rmsequad  2370.954357\n",
              "2       rmseexpo  1620.925946\n",
              "3        rmseadd  6902.612573\n",
              "4  rmseaddlinear  1303.766907\n",
              "5    rmseaddquad  2598.575352\n",
              "6      rmsemulin  1652.464095\n",
              "7  rmse_mul_quad  2419.854410"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KxjeTIV-IHQE"
      },
      "source": [
        "pred=pd.read_excel(\"/content/requiredforprediction.xlsx\",index_col=0)"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 195
        },
        "id": "P2Ph3OueIQPN",
        "outputId": "038a345f-32b4-41be-9f93-76a1393d669a"
      },
      "source": [
        "pred.head()"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>quarter</th>\n",
              "      <th>value</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Q1</th>\n",
              "      <th>Q2</th>\n",
              "      <th>Q3</th>\n",
              "      <th>Q4</th>\n",
              "      <th>t</th>\n",
              "      <th>t_square</th>\n",
              "      <th>log_value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>82</th>\n",
              "      <td>2014-Q3</td>\n",
              "      <td>23965.3</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>83</td>\n",
              "      <td>6889</td>\n",
              "      <td>10.084362</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>83</th>\n",
              "      <td>2014-Q4</td>\n",
              "      <td>23474.8</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>84</td>\n",
              "      <td>7056</td>\n",
              "      <td>10.063683</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>84</th>\n",
              "      <td>2015-Q1</td>\n",
              "      <td>23384.8</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>85</td>\n",
              "      <td>7225</td>\n",
              "      <td>10.059842</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>85</th>\n",
              "      <td>2015-Q2</td>\n",
              "      <td>23808.7</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>86</td>\n",
              "      <td>7396</td>\n",
              "      <td>10.077806</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86</th>\n",
              "      <td>2015-Q3</td>\n",
              "      <td>24311.7</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>87</td>\n",
              "      <td>7569</td>\n",
              "      <td>10.098713</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    quarter    value Quarter  Q1  Q2  Q3  Q4   t  t_square  log_value\n",
              "82  2014-Q3  23965.3      Q3   0   0   1   0  83      6889  10.084362\n",
              "83  2014-Q4  23474.8      Q4   0   0   0   1  84      7056  10.063683\n",
              "84  2015-Q1  23384.8      Q1   1   0   0   0  85      7225  10.059842\n",
              "85  2015-Q2  23808.7      Q2   0   1   0   0  86      7396  10.077806\n",
              "86  2015-Q3  24311.7      Q3   0   0   1   0  87      7569  10.098713"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 55
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "DJwUYszWL_nC"
      },
      "source": [
        "Linear model is choosen as good fit for prediction"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d20DIIS7FLP2"
      },
      "source": [
        "final_pred=smf.ols('value~t',data=df1).fit()"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 639
        },
        "id": "w-lgFsn9Fixv",
        "outputId": "3c75636e-f710-41e5-e8ee-27b175e31b1f"
      },
      "source": [
        "pred_new = pd.Series(linear.predict(pred))\n",
        "pred_new\n",
        "\n",
        "pred[\"forecasted_value\"] = pd.Series(pred_new)\n",
        "\n",
        "pred"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>quarter</th>\n",
              "      <th>value</th>\n",
              "      <th>Quarter</th>\n",
              "      <th>Q1</th>\n",
              "      <th>Q2</th>\n",
              "      <th>Q3</th>\n",
              "      <th>Q4</th>\n",
              "      <th>t</th>\n",
              "      <th>t_square</th>\n",
              "      <th>log_value</th>\n",
              "      <th>forecasted_value</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>82</th>\n",
              "      <td>2014-Q3</td>\n",
              "      <td>23965.3</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>83</td>\n",
              "      <td>6889</td>\n",
              "      <td>10.084362</td>\n",
              "      <td>25018.723811</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>83</th>\n",
              "      <td>2014-Q4</td>\n",
              "      <td>23474.8</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>84</td>\n",
              "      <td>7056</td>\n",
              "      <td>10.063683</td>\n",
              "      <td>24905.248393</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>84</th>\n",
              "      <td>2015-Q1</td>\n",
              "      <td>23384.8</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>85</td>\n",
              "      <td>7225</td>\n",
              "      <td>10.059842</td>\n",
              "      <td>24791.772975</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>85</th>\n",
              "      <td>2015-Q2</td>\n",
              "      <td>23808.7</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>86</td>\n",
              "      <td>7396</td>\n",
              "      <td>10.077806</td>\n",
              "      <td>24678.297557</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>86</th>\n",
              "      <td>2015-Q3</td>\n",
              "      <td>24311.7</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>87</td>\n",
              "      <td>7569</td>\n",
              "      <td>10.098713</td>\n",
              "      <td>24564.822139</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>87</th>\n",
              "      <td>2015-Q4</td>\n",
              "      <td>25315.1</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>88</td>\n",
              "      <td>7744</td>\n",
              "      <td>10.139156</td>\n",
              "      <td>24451.346722</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>88</th>\n",
              "      <td>2016-Q1</td>\n",
              "      <td>23129.9</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>89</td>\n",
              "      <td>7921</td>\n",
              "      <td>10.048881</td>\n",
              "      <td>24337.871304</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>89</th>\n",
              "      <td>2016-Q2</td>\n",
              "      <td>23324.9</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>90</td>\n",
              "      <td>8100</td>\n",
              "      <td>10.057277</td>\n",
              "      <td>24224.395886</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>90</th>\n",
              "      <td>2016-Q3</td>\n",
              "      <td>23889.9</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>91</td>\n",
              "      <td>8281</td>\n",
              "      <td>10.081211</td>\n",
              "      <td>24110.920468</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>91</th>\n",
              "      <td>2016-Q4</td>\n",
              "      <td>23206.6</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>92</td>\n",
              "      <td>8464</td>\n",
              "      <td>10.052192</td>\n",
              "      <td>23997.445050</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>92</th>\n",
              "      <td>2017-Q1</td>\n",
              "      <td>21934.5</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>93</td>\n",
              "      <td>8649</td>\n",
              "      <td>9.995816</td>\n",
              "      <td>23883.969633</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>93</th>\n",
              "      <td>2017-Q2</td>\n",
              "      <td>22573.1</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>94</td>\n",
              "      <td>8836</td>\n",
              "      <td>10.024514</td>\n",
              "      <td>23770.494215</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>94</th>\n",
              "      <td>2017-Q3</td>\n",
              "      <td>23162.8</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>95</td>\n",
              "      <td>9025</td>\n",
              "      <td>10.050303</td>\n",
              "      <td>23657.018797</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>95</th>\n",
              "      <td>2017-Q4</td>\n",
              "      <td>23253.4</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>96</td>\n",
              "      <td>9216</td>\n",
              "      <td>10.054207</td>\n",
              "      <td>23543.543379</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>96</th>\n",
              "      <td>2018-Q1</td>\n",
              "      <td>22341.0</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>97</td>\n",
              "      <td>9409</td>\n",
              "      <td>10.014179</td>\n",
              "      <td>23430.067961</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>97</th>\n",
              "      <td>2018-Q2</td>\n",
              "      <td>22628.2</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>98</td>\n",
              "      <td>9604</td>\n",
              "      <td>10.026952</td>\n",
              "      <td>23316.592544</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98</th>\n",
              "      <td>2018-Q3</td>\n",
              "      <td>22552.9</td>\n",
              "      <td>Q3</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>99</td>\n",
              "      <td>9801</td>\n",
              "      <td>10.023619</td>\n",
              "      <td>23203.117126</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>2018-Q4</td>\n",
              "      <td>22504.0</td>\n",
              "      <td>Q4</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>100</td>\n",
              "      <td>10000</td>\n",
              "      <td>10.021448</td>\n",
              "      <td>23089.641708</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>100</th>\n",
              "      <td>2019-Q1</td>\n",
              "      <td>20682.0</td>\n",
              "      <td>Q1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>101</td>\n",
              "      <td>10201</td>\n",
              "      <td>9.937019</td>\n",
              "      <td>22976.166290</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>101</th>\n",
              "      <td>2019-Q2</td>\n",
              "      <td>20491.9</td>\n",
              "      <td>Q2</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>102</td>\n",
              "      <td>10404</td>\n",
              "      <td>9.927785</td>\n",
              "      <td>22862.690872</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "     quarter    value Quarter  Q1  ...    t  t_square  log_value  forecasted_value\n",
              "82   2014-Q3  23965.3      Q3   0  ...   83      6889  10.084362      25018.723811\n",
              "83   2014-Q4  23474.8      Q4   0  ...   84      7056  10.063683      24905.248393\n",
              "84   2015-Q1  23384.8      Q1   1  ...   85      7225  10.059842      24791.772975\n",
              "85   2015-Q2  23808.7      Q2   0  ...   86      7396  10.077806      24678.297557\n",
              "86   2015-Q3  24311.7      Q3   0  ...   87      7569  10.098713      24564.822139\n",
              "87   2015-Q4  25315.1      Q4   0  ...   88      7744  10.139156      24451.346722\n",
              "88   2016-Q1  23129.9      Q1   1  ...   89      7921  10.048881      24337.871304\n",
              "89   2016-Q2  23324.9      Q2   0  ...   90      8100  10.057277      24224.395886\n",
              "90   2016-Q3  23889.9      Q3   0  ...   91      8281  10.081211      24110.920468\n",
              "91   2016-Q4  23206.6      Q4   0  ...   92      8464  10.052192      23997.445050\n",
              "92   2017-Q1  21934.5      Q1   1  ...   93      8649   9.995816      23883.969633\n",
              "93   2017-Q2  22573.1      Q2   0  ...   94      8836  10.024514      23770.494215\n",
              "94   2017-Q3  23162.8      Q3   0  ...   95      9025  10.050303      23657.018797\n",
              "95   2017-Q4  23253.4      Q4   0  ...   96      9216  10.054207      23543.543379\n",
              "96   2018-Q1  22341.0      Q1   1  ...   97      9409  10.014179      23430.067961\n",
              "97   2018-Q2  22628.2      Q2   0  ...   98      9604  10.026952      23316.592544\n",
              "98   2018-Q3  22552.9      Q3   0  ...   99      9801  10.023619      23203.117126\n",
              "99   2018-Q4  22504.0      Q4   0  ...  100     10000  10.021448      23089.641708\n",
              "100  2019-Q1  20682.0      Q1   1  ...  101     10201   9.937019      22976.166290\n",
              "101  2019-Q2  20491.9      Q2   0  ...  102     10404   9.927785      22862.690872\n",
              "\n",
              "[20 rows x 11 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 56
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6l2d6eKYJ1yZ"
      },
      "source": [
        ""
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}