Heterogeneous first-stage long-run coefficients option

[hborrero]

When estimating using the Fully-modified OLS (FMOLS) method, selecting the "pooled" panel method, what is the "heterogeneous first-stage long-run coefficients" option for??? In which cases should it be selected, and in which cases not???

[EViews Glenn]

From the manual...

"Lastly, you can specify the form of the first-stage cointegrating equation regression that EViews uses to obtain [the residuals] for computing the long-run covariances. By default, the first-stage regression assumes homogeneous long-run coefficients, but you may allow for different coefficients by selecting the Heterogeneous first-stage long-run coefficients checkbox."

http://www.eviews.com/help/helpintro.ht ... Views.html

[hborrero]

Thank you very much, Mr. Glenn;

But still, theoretically, what does allowing for "heterogeneous first-stage long-run coefficients" mean? When should it be applied? Why does it generally increments the likelyhood of the model?

Thank you very much in advance for your time.

[EViews Glenn]

When you estimate the first-stage long-run regression you can constrain the coefficients to be the same across all cross-sections or you can estimate a separate long-run regression for each cross-section. Which you should choose depends on whether you wish to perform the estimation using the restriction or not. Obviously the residuals that you will obtain from the latter will be smaller. Whether they are better or not is an empirical question.

[hborrero]

You are very kind, Mr. Glenn. I appreciate very much your help. I wish you great happiness!