Need advices and explanation

For econometric discussions not necessarily related to EViews.

Moderators: EViews Gareth, EViews Moderator

cuongnh
Posts: 70
Joined: Thu Dec 11, 2008 10:09 am
Contact:

Need advices and explanation

Postby cuongnh » Sun May 10, 2009 6:25 am

Dear, I have used 2-step Cochrane-Orcutt to regress my data. In the first OLS, the P-values for all few dummy variable used in the model show good significance. However, after using 2-step C-O (using rho value obtained from the OLS to caculate new data for independent and dependent variable (known as transformation), then new model is OLS again), the p-values of some dummy variable become not significant. I dont know how to explain or interpret this, or whether I should use those dummy variables or not.
And, one more thing relating to C-O process, the step which has minimum sum of squares residual gives the best estimates or results, however sum of squares residual is positively responsive to R^2, it means R^2 of the step with best estimates is smallest. How should I understand R^2 value in this case, is that R^2 low, it explains certain abnormality in the model?
Many thanks!

Return to “Econometric Discussions”

Who is online

Users browsing this forum: No registered users and 21 guests