: Stata will report if your panel is "strongly balanced" (all entities observed for all time periods) or "unbalanced" (some gaps). 2. Exploratory Analysis
Standard summaries mask within-panel variation. Use xtsum : stata panel data
The null hypothesis of the Hausman test is that the preferred model is Random Effects vs. the alternative Fixed Effects. It essentially tests whether the unique errors are correlated with the regressors. How to Run the Hausman Test in Stata : Stata will report if your panel is
Pooled Ordinary Least Squares (OLS) acts as if the panel structure does not exist, simply pooling all observations together. Use xtsum : The null hypothesis of the
– Deciding between FE and RE:
: If your variables are correlated with the error term, use Instrumental Variables ( xtivreg ).
Treats the data as one big cross-section, ignoring the panel structure.