Dear my friend!
I would line to run a panel data for 40 cases over 2012-2016 (4 years). My variables comprise 13 series (6 real and 7 dummy variables). I faced a problem in running a FMOLS which seems the numbers of variables and observations are not proper. I'd be thankful if you had any useful comments for me.
Yours.
Your problem statement is not clear. what do you mean by 40 cases? If you can clearly state your objective, then i may be able to answer your query. Any estimation methodology whether it is fixed effect, random effect, FMOLS, etc, requires some prerequisites and your underlying objective. In the light of your problem statement, i can suggest you the appropriate estimation methodology.
fixed effect model provides constant for each cross-sectional unit in the panel data. in your case, i think you have 40 cross-sectional units on which you collected time series obervations for 4 years.
you have also mentioned that you have 13 independent variables in your model, as the number of observations for each case is more than the number of variables, regression can not produce the results.
one of the preconditions for running the regression analysis is the sample,at least,should be more than the number of variables.
you try random effectmodel, probably it may work.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running