Number of Observations on STATA Unequal

[Kezia Akwaboa]

I am regressing The dependent Variable Wellbeing against a dummy variable ' becoming unemployed' , as well as several control variables I have 29,992 observations. However when I regress using either OLS or Ordered probit , the regressions are only done on 26, 201.
Could anyone inform me of why this is and how I can fix this issue.
I also need to run a Fixed effect Ordered legit regression( FE-OL) regression but I can't seem to find it on STATA under Statistics. Could anyone please inform me of where this command is.

[Clyde Schechter]

In any regression command, any observation that has a missing value for any of the variables in the regression command is omitted. So even if most of your variables have only a small number of missing values, when you throw in a large number of variables, you can end up losing a large number of observations.

Notice that when you use only the becameunemp variable, you come pretty close to your full data set being included. If you run -count if missing(Wellbeing, becameunemp)- that number will account for all of the missing observations in the second regression you show.

The first regression has even fewer observations included because you also lose any observation where there is a missing value on any of the other variables.

The only way to fix it is to get values for the missing data. Usually that is infeasible.

[Kezia Akwaboa]

Thank you so much for the help.
I have a query in regards to suing estimators in STATA. I would appreciate any advice you have .

I am regressing The dependent Variable Wellbeing against a dummy variable ' becoming unemployed' , as well as several control variables I have 29,992 observations. However when I regress using either OLS or Ordered probit , the regressions are only done on 26, 201.
Could anyone inform me of why this is and how I can fix this issue.
I also need to run a Fixed effect Ordered legit regression( FE-OL) regression but I can't seem to find it on STATA under Statistics. Could anyone please inform me of where this command is.

[Marcos Almeida]

Please read #2.

You may consider multiple imputation when dealing with missing data.