Statistical programming, matrix languages, and more

question about missing data

Not applicable
Posts: 0

question about missing data

I was tasked to do a panel analysis but unsure how to go forward. I have already obtained the dataset for 112 countries for periods 1980-2005. However, there are some countries/years with missing values. I wanted to use the tscsreg canned procedure so I don't have to program a code [since I'm also not an expert on this]. I just want to do some simple regression analysis.

I learned that there are so many ways to deal with missing data but i'm not quite familiar with analyzing panel data. I read that you can put/treat missing values as "zero" but if I do that it will complicate with my other variables which values are really "zero". When I replaced the missing values with zero, it ran and gave me some results but unsure whether those results are no bias.

really needs help on this one.

thanks for your help.
Posts: 4,181

Re: question about missing data

Posted in reply to deleted_user
I also am not an expert in this area. Try posting
this question to the Time Series and Forecasting Forum:
I'll bet somewhere there can advise.
Ask a Question
Discussion stats
  • 1 reply
  • 2 in conversation