cem command error

[Sara Urbina]

Hello,

I have panel data for a sample for program participants and want to check for balance between the treatment and control groups. imb command, found here, by downloading the cem command.

All the variables I am using are dummies, lag0-2 have possible values of "." ,1,0, and id_off has possible values between 1-10. When i run the following code

imb lag0 lag1 lag2 id_off, tr(audited)

I obtain the following message:

  L1meas():  3301  subscript invalid
             imbalance():     -  function returned error
                 <istmt>:     -  function returned error

I found this response online, that suggested using fewer dummies.

It indeed works if I run the following codes separatly, and only in the specific combination.

imb lag0 id_off, tr(audited)

imb lag1, tr(audited)

imb lag2 ,tr(audited)

Am I am interested in the overall model level of imbalance I would like to run them together. Do you have any suggestions on how to solve the problem?

Thank you in advance,
Sara

[Phil Bromiley]

You didn't get a quick answer. You're using a user-written program which few of us may be using. There are a bunch of treatment estimators available in Stata 14. Depending on what you want to do, these may be easier (and less likely to have bugs or other problems).