Rolling Regression Estimation in Python dataframe

Question

I have a dataframe like this:

           Date         Y         X1         X2          X3
22   2004-05-12  9.348158e-09  0.000081  0.000028     0.000036   
23   2004-05-13  9.285989e-09  0.000073  0.000081     0.000097   
24   2004-05-14  9.732308e-09  0.000085  0.000073     0.000096   
25   2004-05-17  2.235977e-08  0.000089  0.000085     0.000099   
26   2004-05-18  2.792661e-09  0.000034  0.000089     0.000150   
27   2004-05-19  9.745323e-09  0.000048  0.000034     0.000053 

......

1000   2004-05-20  1.835462e-09  0.000034  0.000048     0.000099   
1001   2004-05-21  3.529089e-09  0.000037  0.000034     0.000043   
1002   2004-05-24  3.453047e-09  0.000043  0.000037     0.000059   
1003   2004-05-25  2.963131e-09  0.000038  0.000043     0.000059   
1004   2004-05-26  1.390032e-09  0.000029  0.000038     0.000054   

I want to run a rolling 100-day window OLS regression estimation, which is:

First for the 101st row, I run a regression of Y-X1,X2,X3 using the 1st to 100th rows, and estimate Y for the 101st row;

Then for the 102nd row, I run a regression of Y-X1,X2,X3 using the 2nd to 101st rows, and estimate Y for the 102nd row;

Then for the 103rd row, I run a regression of Y-X1,X2,X3 using the 2nd to 101st rows, and estimate Y for the 103rd row;

......

Until the last row.

How to do this？

Answer 1

model = pd.stats.ols.MovingOLS(y=df.Y, x=df[['X1', 'X2', 'X3']], 
                               window_type='rolling', window=100, intercept=True)
df['Y_hat'] = model.y_predict

Answer 2

statsmodels 0.11.0 added RollingOLS (Jan2020)

from statsmodels.regression.rolling import RollingOLS

#add constant column to regress with intercept
df['const'] = 1

#fit
model = RollingOLS(endog =df['Y'].values , exog=df[['const','X1','X2','X3']],window=20)
rres = model.fit()
rres.params.tail() #look at last few intercept and coef

Or use R-style regression formula

model = RollingOLS.from_formula('Y ~ X1 + X2 + X3' , data = df, window=20)
rres = model.fit()
rres.params.tail()