From c06e5bad456ba9f5a82a4e8f2b04606971fcf146 Mon Sep 17 00:00:00 2001
From: Tetsu Haruyama <tetsu.yes@gmail.com>
Date: Sun, 21 Jul 2024 14:15:34 +0900
Subject: [PATCH] ols -> smf.ols : 1

---
 10_Residuals.ipynb          | 22 +++++++++++-----------
 11_Inference.ipynb          |  8 ++++----
 12_Asymptotics.ipynb        | 14 +++++++-------
 13_Dummies.ipynb            | 26 +++++++++++++-------------
 14_Hetero.ipynb             | 14 +++++++-------
 15_Pooling.ipynb            | 19 ++++++++++---------
 16_linearmodels.ipynb       |  9 ++++++---
 8_Simple_Regression.ipynb   | 14 +++++++-------
 9_Multiple_Regression.ipynb | 16 ++++++++--------
 9 files changed, 73 insertions(+), 69 deletions(-)

diff --git a/10_Residuals.ipynb b/10_Residuals.ipynb
index 99753249..1e8f26f4 100644
--- a/10_Residuals.ipynb
+++ b/10_Residuals.ipynb
@@ -27,8 +27,8 @@
     "import numpy as np\n",
     "import pandas as pd\n",
     "import statsmodels.api as sm\n",
+    "import statsmodels.formula.api as smf\n",
     "\n",
-    "from statsmodels.formula.api import ols\n",
     "from scipy.stats import norm, uniform\n",
     "\n",
     "# 警告メッセージを非表示\n",
@@ -131,7 +131,7 @@
     "    \n",
     "    df = pd.DataFrame({'Y':y, 'X':x})  # DataFrame\n",
     "    \n",
-    "    res = ols(formula='Y ~ X', data=df).fit()  # OLSの計算\n",
+    "    res = smf.ols(formula='Y ~ X', data=df).fit()  # OLSの計算\n",
     "    u_standardized = res.get_influence().resid_studentized_internal  # 標準化残差\n",
     "    \n",
     "    return x, y, res.fittedvalues, res.resid, u_standardized, res.rsquared  # 返り値の設定"
@@ -596,7 +596,7 @@
     "\n",
     "df_diag = pd.DataFrame({'Y':y, 'X':x})  # DataFrameの作成\n",
     "\n",
-    "res_diag = ols(formula='Y ~ X', data=df_diag).fit()  # OLS推定"
+    "res_diag = smf.ols(formula='Y ~ X', data=df_diag).fit()  # OLS推定"
    ]
   },
   {
@@ -914,7 +914,7 @@
     "\n",
     "df = pd.DataFrame({'Y':y, 'X':x})  # DataFrame\n",
     "\n",
-    "res = ols(formula='Y ~ X', data=df).fit()  # OLSの計算\n",
+    "res = smf.ols(formula='Y ~ X', data=df).fit()  # OLSの計算\n",
     "resid_std = res.get_influence().resid_studentized_internal  # 標準化残差\n",
     "\n",
     "plt.scatter(res.fittedvalues,resid_std)  # 散布図\n",
@@ -991,7 +991,7 @@
     "y = 1.0 + 0.1*x +0.1*x2+ u\n",
     "df = pd.DataFrame({'Y':y, 'X':x})\n",
     "\n",
-    "res = ols(formula='Y ~ X', data=df).fit()\n",
+    "res = smf.ols(formula='Y ~ X', data=df).fit()\n",
     "resid_std = res.get_influence().resid_studentized_internal\n",
     "\n",
     "plt.scatter(res.fittedvalues,resid_std)\n",
@@ -1078,7 +1078,7 @@
     "\n",
     "df = pd.DataFrame({'Y':y, 'X':x})\n",
     "\n",
-    "res = ols(formula='Y ~ X', data=df).fit()\n",
+    "res = smf.ols(formula='Y ~ X', data=df).fit()\n",
     "resid_std = res.get_influence().resid_studentized_internal\n",
     "\n",
     "plt.scatter(res.fittedvalues,resid_std)\n",
@@ -1160,7 +1160,7 @@
     "y = 1 + 0.1*x + x**0.6*u\n",
     "df = pd.DataFrame({'Y':y, 'X':x})\n",
     "\n",
-    "res = ols(formula='Y ~ X', data=df).fit()\n",
+    "res = smf.ols(formula='Y ~ X', data=df).fit()\n",
     "resid_std = res.get_influence().resid_studentized_internal\n",
     "\n",
     "plt.scatter(res.fittedvalues,resid_std)\n",
@@ -1232,10 +1232,10 @@
     "df_cd = pd.DataFrame({'Y':y, 'X':x})\n",
     "\n",
     "# 外れ値がない場合のOLS\n",
-    "res_no = ols(formula='Y ~ X', data=df_cd.loc[0:19,:]).fit()\n",
+    "res_no = smf.ols(formula='Y ~ X', data=df_cd.loc[0:19,:]).fit()\n",
     "\n",
     "# 外れ値がある場合のOLS\n",
-    "res_cd = ols(formula='Y ~ X', data=df_cd).fit()"
+    "res_cd = smf.ols(formula='Y ~ X', data=df_cd).fit()"
    ]
   },
   {
@@ -1348,7 +1348,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.9"
   },
   "nteract": {
    "version": "0.23.1"
@@ -1368,5 +1368,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }
diff --git a/11_Inference.ipynb b/11_Inference.ipynb
index bde3f3c6..4a20ff08 100644
--- a/11_Inference.ipynb
+++ b/11_Inference.ipynb
@@ -25,9 +25,9 @@
     "import matplotlib.pyplot as plt\n",
     "import numpy as np\n",
     "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
     "import wooldridge\n",
     "\n",
-    "from statsmodels.formula.api import ols\n",
     "from scipy.stats import t, f\n",
     "\n",
     "# 警告メッセージを非表示\n",
@@ -505,7 +505,7 @@
    "outputs": [],
    "source": [
     "formula_gpa = 'colGPA ~ hsGPA + ACT + skipped'\n",
-    "res_gpa = ols(formula_gpa, data=gpa).fit()"
+    "res_gpa = smf.ols(formula_gpa, data=gpa).fit()"
    ]
   },
   {
@@ -988,7 +988,7 @@
    "outputs": [],
    "source": [
     "formula_0 = 'np.log(salary) ~ years + gamesyr + bavg + hrunsyr + rbisyr'\n",
-    "res_0 = ols(formula_0, data=mlb1).fit()"
+    "res_0 = smf.ols(formula_0, data=mlb1).fit()"
    ]
   },
   {
@@ -1033,7 +1033,7 @@
    "outputs": [],
    "source": [
     "formula_1 = 'np.log(salary) ~ years + gamesyr'\n",
-    "res_1 = ols(formula_1, data=mlb1).fit()"
+    "res_1 = smf.ols(formula_1, data=mlb1).fit()"
    ]
   },
   {
diff --git a/12_Asymptotics.ipynb b/12_Asymptotics.ipynb
index cb7c91f6..58d4f73d 100644
--- a/12_Asymptotics.ipynb
+++ b/12_Asymptotics.ipynb
@@ -25,10 +25,10 @@
     "import matplotlib.pyplot as plt\n",
     "import numpy as np\n",
     "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
     "import wooldridge\n",
     "\n",
     "from scipy.stats import gaussian_kde, t\n",
-    "from statsmodels.formula.api import ols\n",
     "from statsmodels.api import qqplot\n",
     "from statsmodels.stats.stattools import jarque_bera, omni_normtest\n",
     "from numba import njit\n",
@@ -1236,7 +1236,7 @@
    "source": [
     "wage1 = wooldridge.data('wage1')\n",
     "formula_wage = 'wage ~ educ + exper+ tenure'\n",
-    "res_wage = ols(formula_wage, data=wage1).fit()\n",
+    "res_wage = smf.ols(formula_wage, data=wage1).fit()\n",
     "qqplot(res_wage.resid, line='45',fit=True)\n",
     "pass"
    ]
@@ -1260,7 +1260,7 @@
    "source": [
     "wage1 = wooldridge.data('wage1')\n",
     "formula_wage_log = 'np.log(wage) ~ educ + exper+ tenure'\n",
-    "res_wage_log = ols(formula_wage_log, data=wage1).fit()\n",
+    "res_wage_log = smf.ols(formula_wage_log, data=wage1).fit()\n",
     "qqplot(res_wage_log.resid, line='45',fit=True)\n",
     "pass"
    ]
@@ -1613,7 +1613,7 @@
    "outputs": [],
    "source": [
     "form_0 = 'narr86 ~ pcnv + ptime86 + qemp86 + avgsen + tottime'\n",
-    "res_0 = ols(form_0, data=crime1).fit()\n",
+    "res_0 = smf.ols(form_0, data=crime1).fit()\n",
     "res_0.params"
    ]
   },
@@ -1646,7 +1646,7 @@
    "outputs": [],
    "source": [
     "form_1 = 'narr86 ~ pcnv + ptime86 + qemp86'\n",
-    "res_1 = ols(form_1, data=crime1).fit()\n",
+    "res_1 = smf.ols(form_1, data=crime1).fit()\n",
     "res_1.params"
    ]
   },
@@ -1710,7 +1710,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.9"
   },
   "nteract": {
    "version": "0.15.0"
@@ -1730,5 +1730,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }
diff --git a/13_Dummies.ipynb b/13_Dummies.ipynb
index 9d090317..43c2a2a7 100644
--- a/13_Dummies.ipynb
+++ b/13_Dummies.ipynb
@@ -24,8 +24,8 @@
    "source": [
     "import numpy as np\n",
     "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
     "import wooldridge\n",
-    "from statsmodels.formula.api import ols\n",
     "\n",
     "# 警告メッセージを非表示\n",
     "import warnings\n",
@@ -114,7 +114,7 @@
    "source": [
     "form_const = 'wage ~ 1'  # 定数項だけの場合は１が必要\n",
     "\n",
-    "res_const = ols(form_const, data=wage1).fit()\n",
+    "res_const = smf.ols(form_const, data=wage1).fit()\n",
     "\n",
     "res_const.params"
    ]
@@ -213,7 +213,7 @@
    "source": [
     "form_const_2 = 'wage ~ female'\n",
     "\n",
-    "res_const_2 = ols(form_const_2, data=wage1).fit()\n",
+    "res_const_2 = smf.ols(form_const_2, data=wage1).fit()\n",
     "\n",
     "res_const_2.params"
    ]
@@ -481,7 +481,7 @@
    "source": [
     "form_const_4 = 'wage ~ marmale + singfem + marfem'\n",
     "\n",
-    "res_const_4 = ols(form_const_4, data=wage1).fit()\n",
+    "res_const_4 = smf.ols(form_const_4, data=wage1).fit()\n",
     "\n",
     "para4 = res_const_4.params\n",
     "para4"
@@ -631,7 +631,7 @@
    "source": [
     "form_1 = 'wage ~ female + educ + exper+ tenure'\n",
     "\n",
-    "res_1 = ols(form_1, data=wage1).fit()\n",
+    "res_1 = smf.ols(form_1, data=wage1).fit()\n",
     "\n",
     "res_1.params"
    ]
@@ -712,7 +712,7 @@
    "source": [
     "form_2 = 'np.log(wage) ~ female + female:educ + educ + exper + tenure'\n",
     "\n",
-    "res_2 = ols(form_2, data=wage1).fit()"
+    "res_2 = smf.ols(form_2, data=wage1).fit()"
    ]
   },
   {
@@ -891,7 +891,7 @@
    "source": [
     "form_c = 'wage ~  C(sex) + educ'\n",
     "\n",
-    "res_c = ols(form_c, data=df).fit()\n",
+    "res_c = smf.ols(form_c, data=df).fit()\n",
     "\n",
     "res_c.params"
    ]
@@ -936,7 +936,7 @@
    "source": [
     "form_cm = 'wage ~  C(sex,Treatment(\"male\")) + educ'\n",
     "\n",
-    "res_cm = ols(form_cm, data=df).fit()\n",
+    "res_cm = smf.ols(form_cm, data=df).fit()\n",
     "\n",
     "res_cm.params"
    ]
@@ -960,7 +960,7 @@
    "source": [
     "form_ca = 'wage ~  female + educ'\n",
     "\n",
-    "res_ca = ols(form_ca, data=df).fit()\n",
+    "res_ca = smf.ols(form_ca, data=df).fit()\n",
     "\n",
     "res_ca.params"
    ]
@@ -1046,7 +1046,7 @@
    "source": [
     "form_p = 'np.log(pfries) ~ prpblck + np.log(income) + C(chain)'\n",
     "\n",
-    "res_p = ols(form_p, data=discrim).fit()\n",
+    "res_p = smf.ols(form_p, data=discrim).fit()\n",
     "\n",
     "print(res_p.summary().tables[1])"
    ]
@@ -1202,7 +1202,7 @@
    "source": [
     "form_c = 'np.log(pfries) ~ prpblck + np.log(income) + chain'\n",
     "\n",
-    "res_c = ols(form_c, data=df_c).fit()\n",
+    "res_c = smf.ols(form_c, data=df_c).fit()\n",
     "\n",
     "print(res_c.summary().tables[1])"
    ]
@@ -1237,7 +1237,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.9"
   },
   "nteract": {
    "version": "0.15.0"
@@ -1257,5 +1257,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }
diff --git a/14_Hetero.ipynb b/14_Hetero.ipynb
index 47b33811..dc299ee1 100644
--- a/14_Hetero.ipynb
+++ b/14_Hetero.ipynb
@@ -27,10 +27,10 @@
     "import lmdiag\n",
     "import numpy as np\n",
     "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
     "import wooldridge\n",
     "\n",
     "from seaborn import residplot\n",
-    "from statsmodels.formula.api import ols\n",
     "from statsmodels.stats.api import het_breuschpagan, het_white\n",
     "from statsmodels.stats.outliers_influence import reset_ramsey\n",
     "\n",
@@ -215,7 +215,7 @@
    "source": [
     "form_ols = 'cumgpa ~ sat + hsperc + tothrs + female + black + white'\n",
     "\n",
-    "mod_ols = ols(form_ols, data=gpa3)\n",
+    "mod_ols = smf.ols(form_ols, data=gpa3)\n",
     "res_ols = mod_ols.fit()\n",
     "\n",
     "print(res_ols.summary().tables[1])"
@@ -328,7 +328,7 @@
    },
    "outputs": [],
    "source": [
-    "res_HC3 = ols(form_ols, data=gpa3).fit(cov_type='HC3', use_t=True)\n",
+    "res_HC3 = smf.ols(form_ols, data=gpa3).fit(cov_type='HC3', use_t=True)\n",
     "\n",
     "print(res_HC3.summary().tables[1])"
    ]
@@ -522,7 +522,7 @@
    "source": [
     "form_h = 'price ~ lotsize + sqrft + bdrms'\n",
     "\n",
-    "res_h = ols(form_h, data=hprice1).fit()\n",
+    "res_h = smf.ols(form_h, data=hprice1).fit()\n",
     "\n",
     "print(res_h.summary().tables[1])"
    ]
@@ -595,7 +595,7 @@
    "source": [
     "form_h_log = 'np.log(price) ~ np.log(lotsize) + np.log(sqrft) + bdrms'\n",
     "\n",
-    "res_h_log = ols(form_h_log, data=hprice1).fit()\n",
+    "res_h_log = smf.ols(form_h_log, data=hprice1).fit()\n",
     "\n",
     "print(res_h_log.summary().tables[1])"
    ]
@@ -1090,7 +1090,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.9"
   },
   "nteract": {
    "version": "0.15.0"
@@ -1110,5 +1110,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }
diff --git a/15_Pooling.ipynb b/15_Pooling.ipynb
index 21a73458..81ffbdfd 100644
--- a/15_Pooling.ipynb
+++ b/15_Pooling.ipynb
@@ -24,9 +24,10 @@
    "source": [
     "import numpy as np\n",
     "import pandas as pd\n",
+    "import statsmodels.formula.api as smf\n",
     "import wooldridge\n",
+    "\n",
     "from scipy.stats import t\n",
-    "from statsmodels.formula.api import ols\n",
     "\n",
     "# 警告メッセージを非表示\n",
     "import warnings\n",
@@ -187,7 +188,7 @@
     "                   y85:educ + y85:female + \\\n",
     "                   exper + I((exper**2)/100) + union'\n",
     "\n",
-    "result = ols(formula, cps).fit()"
+    "result = smf.ols(formula, cps).fit()"
    ]
   },
   {
@@ -489,7 +490,7 @@
    "source": [
     "formula = 'rprice ~ nearinc + y81 + nearinc:y81'\n",
     "\n",
-    "result = ols(formula, data=kielmc).fit()"
+    "result = smf.ols(formula, data=kielmc).fit()"
    ]
   },
   {
@@ -566,7 +567,7 @@
    "source": [
     "formula_1 = 'np.log(rprice) ~ nearinc * y81'\n",
     "\n",
-    "result_1 = ols(formula_1, data=kielmc).fit()\n",
+    "result_1 = smf.ols(formula_1, data=kielmc).fit()\n",
     "\n",
     "print(result_1.summary().tables[1])"
    ]
@@ -594,7 +595,7 @@
     "formula_2 = 'np.log(rprice) ~ nearinc * y81 + age + I(age**2) + \\\n",
     "            np.log(intst) + np.log(land) + np.log(area) + rooms + baths'\n",
     "\n",
-    "result_2 = ols(formula_2, data=kielmc).fit()\n",
+    "result_2 = smf.ols(formula_2, data=kielmc).fit()\n",
     "\n",
     "print(result_2.summary().tables[1])"
    ]
@@ -1174,7 +1175,7 @@
    "source": [
     "formula_1 = 'crmrte_diff ~ unem_diff'\n",
     "\n",
-    "result_1 = ols(formula_1, crime2_diff).fit()\n",
+    "result_1 = smf.ols(formula_1, crime2_diff).fit()\n",
     "\n",
     "print(result_1.summary().tables[1])"
    ]
@@ -1240,7 +1241,7 @@
    "source": [
     "formula_ols_1 = 'crmrte ~ d87 + unem'\n",
     "\n",
-    "result_ols_1 = ols(formula_ols_1, crime2).fit()\n",
+    "result_ols_1 = smf.ols(formula_ols_1, crime2).fit()\n",
     "\n",
     "print(result_ols_1.summary().tables[1])"
    ]
@@ -1271,7 +1272,7 @@
    "source": [
     "formula_ols_2 = 'crmrte ~ unem'\n",
     "\n",
-    "result_ols_2 = ols(formula_ols_2, crime2).fit()\n",
+    "result_ols_2 = smf.ols(formula_ols_2, crime2).fit()\n",
     "\n",
     "print(result_ols_2.summary().tables[1])"
    ]
@@ -1398,7 +1399,7 @@
    },
    "outputs": [],
    "source": [
-    "result_2 = ols(formula_2, crime4).fit()\n",
+    "result_2 = smf.ols(formula_2, crime4).fit()\n",
     "\n",
     "print(result_2.summary().tables[1])"
    ]
diff --git a/16_linearmodels.ipynb b/16_linearmodels.ipynb
index 5c7b4f4c..13d76e2d 100644
--- a/16_linearmodels.ipynb
+++ b/16_linearmodels.ipynb
@@ -25,6 +25,7 @@
     "import pandas as pd\n",
     "import py4macro\n",
     "import wooldridge\n",
+    "\n",
     "from linearmodels.panel.data import PanelData\n",
     "from linearmodels.panel import FirstDifferenceOLS\n",
     "\n",
@@ -929,7 +930,9 @@
     "実際に回帰式を書くことにする。使い方は`statsmodels`と似ている。\n",
     "* `FirstDifferenceOLS`モジュールの関数`.from_formula`を使い次のように引数を指定する。\n",
     "\n",
-    "$$\\text{.from_formula}(\\text{回帰式}, \\text{データ})$$\n",
+    "    $$\n",
+    "    \\text{.from\\_formula}\\left(\\text{回帰式}, \\text{データ}\\right)\n",
+    "    $$\n",
     "\n",
     "* 定数項を入れることはできない仕様となっている。\n",
     "* ここでは，以前の推定結果と比べるために，ダミー変数`d82`を追加する。"
@@ -1039,7 +1042,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.9"
   },
   "toc": {
    "base_numbering": 1,
@@ -1056,5 +1059,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }
diff --git a/8_Simple_Regression.ipynb b/8_Simple_Regression.ipynb
index 022b702e..14b073db 100644
--- a/8_Simple_Regression.ipynb
+++ b/8_Simple_Regression.ipynb
@@ -27,10 +27,10 @@
     "import numpy as np\n",
     "import pandas as pd\n",
     "import py4macro    # 属性を調べるsee()関数が含まれる\n",
+    "import statsmodels.formula.api as smf  # 回帰分析のパッケージ\n",
     "import wooldridge  # wooldridgeのデータ・パッケージ\n",
     "\n",
     "from scipy.stats import norm    # 正規分布関数\n",
-    "from statsmodels.formula.api import ols  # 回帰分析のパッケージ\n",
     "from numba import njit  # シミュレーションの計算を高速化する\n",
     "\n",
     "# 警告メッセージを非表示\n",
@@ -578,7 +578,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "mod = ols(formula, data=df)"
+    "mod = smf.ols(formula, data=df)"
    ]
   },
   {
@@ -1068,7 +1068,7 @@
    },
    "outputs": [],
    "source": [
-    "res = ols(formula, data=df).fit()"
+    "res = smf.ols(formula, data=df).fit()"
    ]
   },
   {
@@ -1330,7 +1330,7 @@
    "outputs": [],
    "source": [
     "formula = 'Y ~ X'\n",
-    "res_sim = ols(formula, data=df_sim).fit()"
+    "res_sim = smf.ols(formula, data=df_sim).fit()"
    ]
   },
   {
@@ -1411,7 +1411,7 @@
     "    df_sim = pd.DataFrame({'X':x, 'Y':y})\n",
     "    \n",
     "    formula = 'Y ~ X'\n",
-    "    res_sim = ols(formula, data=df_sim).fit()\n",
+    "    res_sim = smf.ols(formula, data=df_sim).fit()\n",
     "    \n",
     "    return res_sim.params"
    ]
@@ -1835,7 +1835,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.11.9"
   },
   "nteract": {
    "version": "0.14.5"
@@ -1855,5 +1855,5 @@
   }
  },
  "nbformat": 4,
- "nbformat_minor": 2
+ "nbformat_minor": 4
 }
diff --git a/9_Multiple_Regression.ipynb b/9_Multiple_Regression.ipynb
index 872910ab..71cc90e8 100644
--- a/9_Multiple_Regression.ipynb
+++ b/9_Multiple_Regression.ipynb
@@ -27,13 +27,13 @@
     "import numpy as np\n",
     "import pandas as pd\n",
     "import seaborn as sns\n",
+    "import statsmodels.formula.api as smf\n",
     "import py4macro\n",
     "import wooldridge\n",
     "\n",
     "from numba import njit\n",
     "from pandas.plotting import scatter_matrix\n",
     "from scipy.stats import norm, uniform, gaussian_kde, multivariate_normal\n",
-    "from statsmodels.formula.api import ols\n",
     "from statsmodels.stats.outliers_influence import variance_inflation_factor as vif\n",
     "\n",
     "# 警告メッセージを非表示\n",
@@ -182,7 +182,7 @@
    },
    "outputs": [],
    "source": [
-    "res_1 = ols(formula_1, data=wage1).fit()"
+    "res_1 = smf.ols(formula_1, data=wage1).fit()"
    ]
   },
   {
@@ -506,7 +506,7 @@
    },
    "outputs": [],
    "source": [
-    "res_2 = ols(formula_2, data=wage1).fit()"
+    "res_2 = smf.ols(formula_2, data=wage1).fit()"
    ]
   },
   {
@@ -548,7 +548,7 @@
    "outputs": [],
    "source": [
     "formula_3 = 'np.log(wage) ~ educ + tenure + exper + I(exper**2)'\n",
-    "res_3 = ols(formula_3, data=wage1).fit()"
+    "res_3 = smf.ols(formula_3, data=wage1).fit()"
    ]
   },
   {
@@ -611,7 +611,7 @@
    "outputs": [],
    "source": [
     "formula_4 = 'np.log(wage) ~ educ + exper + tenure + myfunc(exper)'\n",
-    "res_4 = ols(formula_4, data=wage1).fit()"
+    "res_4 = smf.ols(formula_4, data=wage1).fit()"
    ]
   },
   {
@@ -724,7 +724,7 @@
    "outputs": [],
    "source": [
     "formula_1a = 'wage ~ educ + tenure + exper'\n",
-    "res_1a = ols(formula_1a, data=wage1).fit()\n",
+    "res_1a = smf.ols(formula_1a, data=wage1).fit()\n",
     "res_1a.params"
    ]
   },
@@ -1341,7 +1341,7 @@
    "outputs": [],
    "source": [
     "df_check = pd.DataFrame({'Y':y,'X1':x1,'X2':x2})\n",
-    "res_check = ols('Y ~ X1 + X2', data=df_check).fit()\n",
+    "res_check = smf.ols('Y ~ X1 + X2', data=df_check).fit()\n",
     "res_check.params"
    ]
   },
@@ -2520,7 +2520,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.1"
+   "version": "3.11.9"
   },
   "nteract": {
    "version": "0.15.0"