0
$\begingroup$

I am analyzing a longitudinal panel dataset using OLS. The data spans around 40 years, but for some variables data was unavailable for certain categories. In most cases, the data for given category stretches not as far back as for others (or ends earlier), but there are also a few cases where a category has an entire blank series. Unfortunately there is no way that I can retrieve the missing data. Hence, I will have to estimate my models with gaps in my dataset.

I am wondering whether there are consequences for my estimation results that I should be aware of. I'm thinking here statistical biases or issues regarding the validity of my results. Suggestions would be very much appreciated.

$\endgroup$
1
$\begingroup$

Check Missing Data in Longitudinal Studies: A Review presentation. There are books on this subject. It's a very broad subject. I think that MLE is more common approach when dealing with missing data than OLS though.

$\endgroup$

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.