Hi
I'm estimating a model for the freight rate as the dependent variable. My model has three explanatory variables, being two invariants in time (continuous) for each cross section and a dummy referring to seasonality. My database has N = 17 and T = 12 (monthly observations). The Chow, LM / Breusch-Pagan and Hausman tests indicated a random effects model. However, I would like to know how to test and treat heteroskedasticity and other problems such as autocorrelation and cross dependency in the context of a random effects model in Stata. Moreover, I wonder if the random effects estimator is adequate and consistent for a sample with small N and T's.
Related Posts with Heteroskedasticity and autocorrelation in a random effects model with small sample
How do do multilevel multinomial regression for multiply imputed data?- Simulating data for logistic regression with categorical variables
- SYS-GMM: Problem of overidentifying restrictions
- Generate a new variable based on character ("/") of another variable
- (Easy) Quadratic term of variable in regression: Joint hypothesis test
Subscribe to:
Post Comments (Atom)
0 Response to Heteroskedasticity and autocorrelation in a random effects model with small sample
Post a Comment