LASSO with mi estimate:

[Leo Martinez]

Hello,

I am fairly new to Stata. I use Stata 14.1.

I wish to perform LASSO with my data and was able to do this with my data with the following code using the LARS program:

lars prevdis agegroupnew tbpast hiv cough nutritionstatusnew familysize cavtindx windcat relativenew bcg1 indagenew if indagenew<10 & case<3 & windcat<10  & hiv<10 & tbpast<10 & agegroupnew<10 & cough<10 & cavtindx<10 & nutritionstatusnew<10, a(lasso) g

However when I used multiple imputation to take care of missing value this command no longer worked. See below:

mi set wide
mi register impute hiv cough cavtindx relativenew tbpast prevdis
mi describe
set seed 8945
mi impute chained (logit) cough (mlogit) hiv tbpast windcat cavtindx = prevdis familysize indagenew if case<3, add(10)

mi estimate: lars price weight length mpg turn rep78 headroom trunk displacement gear_ratio foreign, a(lasso) g

After this I get the following error:

. mi estimate: lars prevdis agegroupnew tbpast hiv cough nutritionstatusnew familysize cavtindx windcat relativenew bcg1 indagenew if indagenew<10 & case<3 & windcat<10  &
>  hiv<10 & tbpast<10 & agegroupnew<10 & cough<10 & cavtindx<10 & nutritionstatusnew<10, a(lasso) g
mi estimate: command not supported
    lars is not officially supported by mi estimate; see mi estimation for a list of Stata estimation commands that are supported by mi estimate.  You can use option
    cmdok to allow estimation anyway.
r(198);

I attempted to use the cmdok command with mi estimate in order to force the situation and it again gives an error.

. mi estimate, cmdok: lars prevdis agegroupnew tbpast hiv cough nutritionstatusnew familysize cavtindx windcat relativenew bcg1 indagenew if indagenew<10 & case<3 & windca
> t<10  & hiv<10 & tbpast<10 & agegroupnew<10 & cough<10 & cavtindx<10 & nutritionstatusnew<10, a(lasso) g
macro e(cmd) is not set
matrix e(b) is not set
matrix e(V) is not set
r(301);

Is there a way to perform LASSO with multiple imputation on Stata?

Thank you for any help you can provide!

Best,
Leo

[Leo Martinez]

I did some more looking and found this link to a previous question posted in 2009:

http://www.stata.com/statalist/archi.../msg00555.html

They seem to state the same problem that I am having.