數據分析模型之多元回歸Python代碼

多元回歸

  # 導⼊模塊
  from sklearn import model_selection
  # 導⼊數據
  Profit = pd.read_excel(r'Predict to Profit.xlsx')
  # 將數據集拆分為訓練集和測試集
  train, test = model_selection.train_test_split(Profit, 
                                                 test_size = 0.2, 
                                                 random_state=1234
                                                )
  # 根據train數據集建模
  model = sm.formula.ols('Profit ~ RD_Spend+Administration+Marketing_Spend+C(State)', data= train).fit()
  # print('模型的偏回歸系數分別為：\n', model.params)
  # 刪除test數據集中的Profit變量，⽤剩下的⾃變量進⾏預測
  test_X = test.drop(labels = 'Profit', axis = 1)

  pred = model.predict(exog = test_X)
  print('對⽐預測值和實際值的差異：\n',pd.DataFrame({'Prediction':pred,'Real':test.Profit}))

由於地區自變量存在多重共線性，所以系統會自動刪除一個，當然也可以自定義一個

  # ⽣成由State變量衍⽣的啞變量
  dummies = pd.get_dummies(Profit.State)
  # 將啞變量與原始數據集⽔平合並
  Profit_New = pd.concat([Profit,dummies], axis = 1)
  # 刪除State變量和California變量(因為State變量已被分解為啞變量，New York變量需要作為參照組)
  Profit_New.drop(labels = ['State','New York'], axis = 1, inplace = True)
  # 拆分數據集Profit_New
  train, test = model_selection.train_test_split(Profit_New, test_size = 0.2, random_state=1234)
  # 建模
  model2 = sm.formula.ols('Profit~RD_Spend+Administration+Marketing_Spend+Florida+California',
  data = train).fit()
  print('模型的偏回歸系數分別為:\n', model2.params)