1
$\begingroup$

I have an unbalanced panel dataset of around 3000 firms for 5 years. My goal is to see if female-owned firm are less likely to survive, so I have a logit model with female-owned dummy as the key independent variable.

The cross-sectional models for year 2007 and 2009 (if year == 2007, if year == 2009) have significant female-owned variable. However, if I try to run a fixed effect logit across all 5 years then the female-owned coefficient is highly insignificant. I have included the i.year dummies in the panel model, and interact female-owned##year2007 but so far the coefficients are not significant.

So how do I interpret this sort of results? Thank you in advance for your help!

Improve this question
$\endgroup$
1
  • $\begingroup$ We were in recession between 2007 and 2009. You should control for it in your panel model. $\endgroup$
    – london
    Apr 2 '19 at 20:52
1
$\begingroup$

Note that the fixed effects estimates use only within-firms differences, essentially discarding any information about differences between firms. If predictor variables vary greatly across firms but have little variation over time for each firm, then fixed effects estimates will be imprecise and have large standard errors. So, if there is not enough variation or change in gender ownership, I may expect a non significant estimate in fixed effects compared to a cross-sectional model.

Improve this answer
$\endgroup$

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.