I am doing a series of multilevel analysis to identify the availability of diagnostic tests in primary care surgeries in different countries. Surgeries are level 1 and country level 2.
I would like to do it with MCMC (using the -bayes- command), because the sample size for my level 2 is low (19 countries).
My data is survey data and I would like to add a sampling weight
However, when I use -bayes-
bayes, mcmcsize(10000) burnin(5000): melogit crp_available_for_use ib2.public_private2 ib2.group_solo2 centered_distance_lab99 centered_pc_turna_time3 i.bloods_taker3 [pweight=scaled_pweight] || country:
pweights not allowed with the bayes prefix
Thank you
Manuel
0 Response to Sampling weights in bayesian analysis
Post a Comment