Hi all,

I am running a 3-class latent class model on 17 categorical indicators and keep getting error messages. The total sample size is 2771.

Initially I ran the following codes:
gsem (gn10_category gn11_category gn12_category gn13_category gn16_category gn18_category gn19_category gn20_category gn21_category gn22_category gn23_category gn25_category gn27_category gn39_category gn40_category gn41_category gn44_category <-) if complete == 1, ologit lclass(C 3)
And got error message "initial value not feasible". Then I went back to Stata manuscript (intro 12 — Convergence problems and how to solve them) and learned to use -startvalues()- option to assign different method to obtain initial values. Then I typed the following codes:
gsem (gn10_category gn11_category gn12_category gn13_category gn16_category gn18_category gn19_category gn20_category gn21_category gn22_category gn23_category gn25_category gn27_category gn39_category gn40_category gn41_category gn44_category <-) if complete == 1, ologit lclass(C 3) startvalues(iv)
But I got the error message "option startvalues() invalid; method iv not allowed". I could not figure out grammar errors in my codes. I greatly appreciate your any provided clues to help resolve the issue.

Many thanks in advance.

Mengmeng