Consumer Research: Hierarchical Bayes: standard errors
Feb 7, 2020 12:10 PM(130 views)
I am currently using JMP Pro 14.0.0 for analyzing different mixed logit models on my data from a discrete choice experiment. I am using the 'Hierarchical Bayes estimation' under the 'Consumer Platform'. I have 2 questions concerning the Hierarchical Bayes estimation.
i) In the ouput, only the posterior estimate, posterior standard deviation and the confidence interval limits are provided (based on the 2.5 and 97.5 quantiles of the posterior distribution according to the help menu). Does someone know how to calculate the standard error out of these equal-tailed confidence intervals in order to perform a backward selection of my model? I cannot find this information in the users guide. I provided an example of the ouput in attachment.
ii) I also read that the standard deviation of the posterior distribution is the same as the standard error. Why is that?