I am quite new to eviews/ economic analysis and I have been trying to estimate a model for my thesis.
I got a panel structure with 14 cross-sections and 148 time periods (weekly data) per cross-section (N=14, t=148). My dependent variable is the log of sales which should be explained by the ~15 independent variables (as for example the log of advertising spending, distribution and the number of working days).
I figured that with panel data I should either compute a fixed effects or a random effects model.
However, I have a few questions:
- Is there an easy way to check the regression assumptions for the two models via eviews?
- Furthermore, can I use "number of working days" as a independent variable in the fixed effects model (since working days are the same for all cross sections)?
- And last but not least can I use the lag dependent variable (sales (-1)) as a independent variable using a fixed effects model or would I violate any important assumptions doing so?
Thank you very much for your help.
