Hi all,

First post here, but have been learning a lot from forum posts for a while.

I'm using suest, mysuest and multiple imputations for missing data for a cost-effectiveness study.

suest for complete case analysis:
qui glm costs i.group, family(gamma) link(log) eform
estimates store est1
qui regress QALYs i.group
estimates store est2
suest est1 est2

mysuest for multiple imputations analysis:

cap program drop mysuest
program mysuest, eclass properties(mi)
version 14.0
args model1 model2

qui `model1'
estimates store est1
qui `model2'
estimates store est2
suest est1 est2
estimates drop est1 est2

ereturn local title "Seemingly unrelated estimation"
end

//SUR with MI//
mi estimate: mysuest "glm costs i.group, family(gamma) link(log) eform" "regress QALYs i.group"

Then to compute incremental costs and QALYs, I'm hoping to use:

a) margins, dydx(group) predict(equation(est1_mean)) atmeans
b) margins, dydx(group) predict(equation(est2_mean)) atmeans

I have 2 lines of queries.

Firstly, the results for a) for suest for complete case analysis, seems very small (~ -0.192), when the difference between the groups should be in the magnitude of hundreds of dollars.
I'm not sure how else to obtain the modelled margins in my cost outcome between groups for my incremental costs.

Secondly, when I try to run a) or b) after mysuest for multiple imputations analysis, I get this error:
"last estimates not found"

Any advice would be very much appreciated