Pseudo R2 for Poisson in Multiple Imputation

[Monica Muller]

Hi,

I need to find the Pseudo R2 for a Poisson estimate that was run on a multiple imputed data. Is there a command for that?

Here is the code I used to run the Poisson:

mi estimate, post: poisson dv iv1 iv2 iv3

Thanks

[Rich Goldstein]

see #6 in http://www.statalist.org/forums/foru...ng-mi-estimate
this uses logistic but should be easy to modify for poisson

[Monica Muller]

Thank you very much Rich.

Do you mind explaining the first and second lines in your code? I really appreciate it.
I thought rhs are your IVs and hodc is your DV. I adjusted the code using my variables and replacing logistic with poisson but I received the following error:

option or not allowed with mi estimate: poisson

I paste your original code here:

local rhs "armg2 armg3 tbsaburn20 tbsaburn21"
noi mi estimate, or saving(miest, replace): logistic hodc `rhs', vce(cluster site)
qui mi query
local M=r(M)
scalar r2=0
scalar cstat=0
qui mi xeq 1/`M': logistic hodc `rhs'; scalar r2=r2+e(r2_p); lroc, nog; scalar cstat=cstat+r(area)
scalar r2=r2/`M'
scalar cstat=cstat/`M'
noi di "Pseudo R=squared over imputed data = " r2
noi di "C statistic over imputed data = " cstat

[Rich Goldstein]

the first line just sets up a local macro (to reduce the typing burden when I do this for several different models, e.g.) and the second uses that local to set the right-hand-side (or independent or predictor) variables; see

h macro

and the associated manual entry (there is a link at the top of the help file

[Monica Muller]

Thanks! locals were not the problem. I got errors for saving() option. I dropped it and copied my Poisson estimation:

mi estimate, post: poisson DV `ivs'  `controls' , `options'

Now, it works perfectly. I really appreciate your help. The code is super helpful.

[Rich Goldstein]

you're welcome