I have imputed panel data and currently checking the model to see if it's an appropriate fit. As can be seen in my imputation command I set the data to wide before imputation. I was able to run some diagnostic plots but I am unable to pull simple means. As per Nguyen, Carlin & Lee (2017), I would also like to tabulate summary statistics of the observed and imputed data. When I try running this command, however, I keep getting a message that reads, "missing_infractions6" not found. Do I need to store the data in some particular format? Or do I need to reformat it? I was able to run the graphical displays in the mi set wide format.
Imputation model
HTML Code:
set seed 832016 mi set wide mi register imputed infractions6 contact6 pwas1 pwas2 therapy3 therapy6 oyas mi impute chained (nbreg) infractions6 (regress) contact6 pwas1 pwas2 (logit) therapy3 therapy6(ologit) oyas = infractions3 contact3 pwas0 age miles length race exitr, add(15) force dots augment
HTML Code:
mi estimate: mean bai1 if missing_bai1==1 mi estimate: mean bai2 if missing_bai2==1 mi estimate: mean bdi1 if missing_bdi1==1 mi estimate: mean bdi2 if missing_bdi2==1 mi estimate: mean infractions6 if missing_infrations6==1 mi estimate: mean contact6 if missing_contact6==1 mi estimate: mean pwas1 if missing_pwas1==1 mi estimate: mean pwas2 if missing_pwas2==1 mi estimate: prop therapy3 if missing_therapy3==1 mi estimate: prop therapy6 if missing_therapy6==1 mi estimate: prop oyas if missing_oyas==1
Thank you in advance,
Roxy
0 Response to Problems with MI diagnostics
Post a Comment