I'm trying to run a multivariate hierarchical bayesian model (HBM) for several quality measures using bayesmh from a meta analysis that I've published. The point of doing this is to get better power because HBM should allow leveraging obs across all of the outcome measures instead of just those that are shared across the outcomes. For example, say the number of obs (N) across the outcome measures (Q1 Q2 Q3) is the following:
for Q1, N=4
Q2, N=6
Q3, N=7
PROBLEM: But when I run bayesmh across these 3 outcome measures the model results only make use of the 4 studies that have non missing values across the 3 outcome measures (instead of the 17 obs). I've reviewed the documentation but I don't see any clear way to relax the default setting of casewise deletion so that all 17 obs are included in the model. Other packages like STAN, are able to do this quite easily. But perhaps I've missed something - does anyone have ideas on how to do this?
Many thanks for your help.
W
Related Posts with relaxing casewise deletion in hierarchical bayesian model using bayesmh
Adding Variables Gradually in RegressionsHi, I want to add multiple independent variables in the regressions gradually with the same depende…
Calculation of Kolm–Pollack Index.Hello everybody! I.m trying to calculate the Kolm–Pollack Index of inequality for the distribution …
Hypothesis Testing Regression Coefficients: One-sidedHelpful Link: https://www.stata.com/support/faqs/statistics/one-sided-tests-for-coefficients/ Hell…
OutliersDear All, I have a quick question, Is it necessary to remove outliers from survey data? Would that…
Monthly last observation in a temporal seriesHi, I have a temporal series with the follow variables: idcompany, lastprice and date (dd/mm/yyyy). …
Subscribe to:
Post Comments (Atom)
0 Response to relaxing casewise deletion in hierarchical bayesian model using bayesmh
Post a Comment