Hi,

I am using panel data with 820 companies across 19 years (i.e. 15580 observations) and each observation has 18 financials variables within (eg. Revenue, Leverage etc.) I would like to run a fe panel data model to test the impact of company diversification on firm performance (eg. Return on Sales or return on Assets.) I would like to confirm my understanding that in order for a panel data model to be correct, I need to ensure:
1. All IVs, DVs and CVs are normally distributed
2. There is no multicollinearity between the DVs and CVs

Is my above understanding correct and if so, what is the best way to verify this?