I’m doing a project in school and I was wondering if a friendly spirit could give me some feedback and guidance on my statistical approach
I’m running a regression on an unbalanced panel dataset of EU firms, where the dependent variable is a yearly patent count and the independent variables are a mix of country and firm specific variables.
I estimate a random effects model and run a Hausman test to determine whether random or fixed effects model is the most appropriate. The Hausman test is highly significant, indicating that the fixed effect model is the best.
From here I’m told that "I am supposed to cluster my standard errors by firm". Can this be done by chosen different “Coef Covariance method” in the equation estimation window under panel options? And if that is the case, what is the differences between: White cross-section, white period and white (diagonal). Which one would be the most appropriate in my case, if I want to cluster standard errors by firm?
I hope this makes sense! And thanks in advance!
For econometric discussions not necessarily related to EViews.
1 post • Page 1 of 1
Who is online
Users browsing this forum: Baidu [Spider] and 6 guests