I am an economist interested in looking at panel data on mothers, their husbands and their grandparents to determine the effect of the economic shock of the recession on their self-reported health outcomes.

Initially I had planned to fit fixed-effect models in order to control for fixed individual differences.

However, thinking on this further, as my analysis will consider the effects of economic shocks on health outcomes of all adults in this dataset at baseline and then ten years later, I wonder if family should be included as a random factor in order to account for potential correlations between related individuals.

Thus, should the analyses be undertaken using a linear mixed model? Any insights you could provide on this would be great!

  • $\begingroup$ I would consider clustering at family level to take into account such correlations $\endgroup$ – MauOlivares Feb 4 '17 at 17:04
  • $\begingroup$ 1. Family fixed effects are automatically accounted for by including individual fixed effects (unless family ID is changed within the sample period). Including individual fixed effects would be sufficient. 2. You might want to control for family characteristics (such as family income). 3. The meaning of FE and RE in econometrics is different from that in statistics (in linear mixed effects model). $\endgroup$ – chan1142 Feb 5 '17 at 1:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.