st: using binary_mediation with mim2

[Francesca Pesola <PesolaF@cardiff.ac.uk>]

I need to use the binary_mediation command on my imputed data to estimate
 the mediation direct and indirect effects

I first run the command

mim2: binary_mediation, dv () iv() mv()

But I received the following message:

command binary not recognised by mim; try specifying category option

r(198);


So I used the -combine-catgegory option as follows

mim2, category(combine) est(e(effect)): binary_mediation, dv(varlist) mv(varlist) iv(varlist)

However, this does not produce the expected output

Also, when I include the bootstrap command:

mim2, category(combine) est(_b[_cons]) se(_se[_cons]): bootstrap r(indir_1) r(tot_ind) r(dir_eff) r(tot_eff), /// reps(500): binary_mediation, dv(varlist) mv(varlist) iv(varlist)

I receive the following message which suggests that bootstrapping is not supported:

prefix bootstrap not allowed after mim

Can anyone suggest how to best proceed?

Thanks,
F

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/