r3900 error

[Enas Farag]

Hello,

My dataset has 97,392,383 observations and 94 variables. I am estimating a DiD model and currently testing for parallel trends using eventdd command.

I can't get the eventdd command to work as I keep getting r3900 errors. This is the error message I receive
" Mata is out of memory; the operating system refused to supply
what Mata requested. There is no Mata or Stata setting that
affects this, and so nothing in Mata or Stata to reset in order
to get more memory. You must take up the problem with your
operating system."

This is the memory status of my STATA. I am using STATA 18


Any advice? I also faced the same error message with stackedev command.

[Maarten Buis]

You can probably loose most of the variables. If you are not going to use them in your analysis, then they don't need to be in your data taking up memory. There are probably missing values, and you can remove observations with those. After that you can use compress to see if you can store that data more efficiently without loosing information.

If there are no weights in your data, then you can use contract and use fweights as a way to save memory without loosing information.