Good day house, please i am very new in stata, I have fitted 1PL, 2PL and 3PL logistic regression to a data havest from 35 multiple choice answer examination conducted for 403 students. On trying to adapt my own data to code provide in example 28 and one posted to stata blog i using the bellow code: generate id = _n quietly reshape long Q, i(id) j(item) rename Q y fvset base none id item set seed 10 set matsize 500 bayesmh y =({discrim}*({subj:i.id}-{diff:i.item})), reffects(id) likelihood(logit) prior({subj:i.id}, normal(0,1)) prior({discrim}, lognormal(0,1)) prior({diff:i.item}, normal(0,1)) init({discrim} 1) exclude({subj:i.id}) burnin(1000) saving(sim1pl,replace) I have the following error message: bayesmh y =({discrim}*({subj:i.id}-{diff:i.item})), reffects(id) likelihood(logit) invalid bayesmh specification To fit one of the supported models, you must specify a likelihood model in option likelihood() and a prior distribution in option prior(). To fit your own model, you must specify, for example, the name of your posterior-function evaluator in option evaluator(). See bayesmh and bayesmh evaluators for details. r(198); . . prior({subj:i.id}, normal(0,1)) command prior is unrecognized r(199); . . prior({discrim}, lognormal(0,1)) command prior is unrecognized r(199); . . prior({diff:i.item}, normal(0,1)) command prior is unrecognized r(199); . . . . init({discrim} 1) exclude({subj:i.id}) command init is unrecognized r(199); . . burnin(1000) saving(sim1pl,replace) command burnin is unrecognized r(199); . . . . /* code for bayesian rasch */ / is not a valid command name r(199); . . bayesmh y =({subj:}-{theta:}), evaluator(logit) expression {subj:} is not defined r(111); . . redefine(subj:i.id)redefine(theta:i.item) command redefine is unrecognized r(199); . . prior({subj:i.id}, normal(0,1)) command prior is unrecognized r(199);
0 Response to Bayesian iRT using bayesmh
Post a Comment