Autocorrelation Error Interpretation

[pham anh]

Hello, I have a question on Autocorrelation of OLS model.
So I encountered a autocorrelation error, and use Cochrane - Orcutt Two-step Procedure(CORC) to fix it. And my OLS became GLS. And I have no idea how to interpret the coefficient on my initial X and Y with the new model. I mean something like, if X rise by 1 unit then Y rise by 3 units. Below are my real output

Please help me, thank you so much.
fyi: Y is outputs per hour ( productivity), X1 is wages/ week, X2 is working hours/week

 

[FactChecker]

Is the left hand side of the GLS autocorrelation fixed model correct? Shouldn't that be $Y-0.71*Y(-1)$?
(see https://en.wikipedia.org/wiki/Cochrane–Orcutt_estimation )

And a small correction: 248*(1-0.71) = 71.92

With that in mind, the equation only gives you values for the change of $Y$ based on the change of the $X_i$s. So if you want to apply it to current values of $Y$ and $X_i$s, you would have to include the data from the prior time ( $Y(-1)$ and $X_i(-1)$s ).

 

[BWV]

Why not just use GMM to adjust the t-stats? Its equivalent to gls and the most common method used in econometrics(or at least in finance)?

https://en.m.wikipedia.org/wiki/Generalized_method_of_moments