0
$\begingroup$

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

I plan to fit fixed-effect models in order to control for fixed individual differences.

Any insights you could provide on this would be great!

$\endgroup$
-1
$\begingroup$

Remember the generalization rule: Factors whose effects you would report and interpret are fixed and factors you generalize about are random. So experimental units are mostly random factors. If it is the only random factor, a fixed effect model suffices (in fact, the randomness of this only factor is the reason why you need statistics).

In your case, you seem to have two random factors: The individual effect of each person and the "family effect". You don't want to report the names of the families and the individuals as informative, so the generalizsation rule says it's random.

If you would omit the family effect (allowing a fixed model), you would claim that the health outcomes of family members are independent. Nobody would believe this.

So yes, you need a mixed model. The next question is however, if you want to model the family effect as uniform in your sample or as depending on some covariate.

$\endgroup$
  • $\begingroup$ Sounds like a reasonable approach. $\endgroup$ – Horst Grünbusch Feb 5 '17 at 17:03
0
$\begingroup$

You should run some test to figure out if you need to implement two-way, fixed or random effects. Could be a good idea to run a unit root test as well, depending on your time-interval. Control for heterosced, VIF and all that lala.

If you are including lots of predictors then consider using GMM instead of FE model.

Last advice: check out multilevel models or Bayesian models, personally I prefer these alternatives to econometric longitudinal models

$\endgroup$

Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.