1

First time posting, please excuse any mistakes I made.

I just ran a rolling regression with the model: Y ~ X1 + X2 + X3 + X3 in pandas. I know that model.y_fitted returns the fitted values, which is just X * beta.

But what is y_predict? It returned slightly different values compared to y_fitted. I am very curious what type of calculation it is doing.

Sample Code:

model = pd.ols(y=Y, x=X, intercept=False, window=500)
model.y_fitted
model.y_predict

The above returns slightly different values. Click here screenshot of Dataframe

Thanks for your help.

Gary Liu
  • 11
  • 1
  • According to this [answer](http://stackoverflow.com/a/9966145/5276797), "pandas y_predict and y_fitted should give you the same value". Maybe follow the link to find out more... – IanS Feb 19 '16 at 09:48
  • See this thread which is asking the same question: http://stackoverflow.com/questions/9943848/forecasting-using-pandas-ols – tsando Sep 26 '16 at 14:50

0 Answers0