In [18]:

    

df20 = pd.read_csv("../cars04.csv")


    

In [19]:

    

df2 = df20[["EngineSize","Cylinders","Horsepower","HighwayMPG","Weight","WheelBase","Hybrid","SuggestedRetailPrice"]]
df2.head()


    

    
Out[19]:






  

    

      

      
EngineSize
      
Cylinders
      
Horsepower
      
HighwayMPG
      
Weight
      
WheelBase
      
Hybrid
      
SuggestedRetailPrice
    
  
  

    

      
0
      
1.6
      
4
      
103
      
34
      
2370
      
98
      
0
      
11690
    
    

      
1
      
1.6
      
4
      
103
      
34
      
2348
      
98
      
0
      
12585
    
    

      
2
      
2.2
      
4
      
140
      
37
      
2617
      
104
      
0
      
14610
    
    

      
3
      
2.2
      
4
      
140
      
37
      
2676
      
104
      
0
      
14810
    
    

      
4
      
2.2
      
4
      
140
      
37
      
2617
      
104
      
0
      
16385
    
  

In [20]:

    

model1 = sm.OLS(df2.ix[:,-1],df2.ix[:,:-1])
result1 = model1.fit()
print(result1.summary())


    

    




                             OLS Regression Results                             
================================================================================
Dep. Variable:     SuggestedRetailPrice   R-squared:                       0.948
Model:                              OLS   Adj. R-squared:                  0.946
Method:                   Least Squares   F-statistic:                     590.1
Date:                  Mon, 13 Jun 2016   Prob (F-statistic):          8.48e-142
Time:                          01:43:20   Log-Likelihood:                -2425.9
No. Observations:                   234   AIC:                             4866.
Df Residuals:                       227   BIC:                             4890.
Df Model:                             7                                         
Covariance Type:              nonrobust                                         
==============================================================================
                 coef    std err          t      P>|t|      [0.025      0.975]
------------------------------------------------------------------------------
EngineSize -4612.4886   1558.366     -2.960      0.003   -7683.202   -1541.775
Cylinders   3136.5695   1000.218      3.136      0.002    1165.670    5107.469
Horsepower   174.1162     16.966     10.263      0.000     140.685     207.547
HighwayMPG   247.5936    187.573      1.320      0.188    -122.014     617.201
Weight        11.8694      2.757      4.305      0.000       6.437      17.302
WheelBase   -519.2678    122.805     -4.228      0.000    -761.250    -277.285
Hybrid      9125.5752   5953.596      1.533      0.127   -2605.804    2.09e+04
==============================================================================
Omnibus:                      112.515   Durbin-Watson:                   1.061
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              735.101
Skew:                           1.779   Prob(JB):                    2.37e-160
Kurtosis:                      10.921   Cond. No.                     3.92e+04
==============================================================================

Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.
[2] The condition number is large, 3.92e+04. This might indicate that there are
strong multicollinearity or other numerical problems.

In [6]:

    

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
scale_df2 = scaler.fit_transform(df2.ix[:,:-1])


    

In [12]:

    

df2_pd = pd.DataFrame(scale_df2)
df2_pd


    

    
Out[12]:






  

    

      

      
0
      
1
      
2
      
3
      
4
      
5
      
6
    
  
  

    

      
0
      
-1.406789
      
-1.033283
      
-1.514918
      
0.858604
      
-1.793629
      
-1.569017
      
-0.113961
    
  

In [21]:

    

df2_pd.columns = df2.columns[:-1].values


    

In [23]:

    

df2_pd["SuggestedRetailPrice"] = df2.SuggestedRetailPrice


    

In [30]:

    

df2_pd = sm.add_constant(df2_pd)


    

In [31]:

    

model2 = sm.OLS(df2_pd.ix[:,-1],df2_pd.ix[:,:-1])
result2 = model2.fit()
print(result2.summary())


    

    




                             OLS Regression Results                             
================================================================================
Dep. Variable:     SuggestedRetailPrice   R-squared:                       0.782
Model:                              OLS   Adj. R-squared:                  0.775
Method:                   Least Squares   F-statistic:                     115.7
Date:                  Mon, 13 Jun 2016   Prob (F-statistic):           4.36e-71
Time:                          01:54:38   Log-Likelihood:                -2416.9
No. Observations:                   234   AIC:                             4850.
Df Residuals:                       226   BIC:                             4877.
Df Model:                             7                                         
Covariance Type:              nonrobust                                         
==============================================================================
                 coef    std err          t      P>|t|      [0.025      0.975]
------------------------------------------------------------------------------
const       2.976e+04    492.458     60.425      0.000    2.88e+04    3.07e+04
EngineSize -6425.0890   1477.698     -4.348      0.000   -9336.918   -3513.260
Cylinders   5233.8697   1423.641      3.676      0.000    2428.561    8039.178
Horsepower  1.148e+04   1048.622     10.950      0.000    9416.148    1.35e+04
HighwayMPG  3419.6888   1086.706      3.147      0.002    1278.317    5561.061
Weight      6263.6418   1397.928      4.481      0.000    3509.002    9018.282
WheelBase    276.5780   1034.517      0.267      0.789   -1761.955    2315.111
Hybrid        48.5726    685.357      0.071      0.944   -1301.934    1399.080
==============================================================================
Omnibus:                      117.077   Durbin-Watson:                   0.994
Prob(Omnibus):                  0.000   Jarque-Bera (JB):              913.845
Skew:                           1.797   Prob(JB):                    3.64e-199
Kurtosis:                      11.990   Cond. No.                         8.60
==============================================================================

Warnings:
[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.

In [32]:

    

sns.pairplot(df2_pd)


    

    
Out[32]:





<seaborn.axisgrid.PairGrid at 0x7ff0edf024d0>