Hi, I can’t figure out how to make inference about latent parameters on new data in a hierarchical model. First I simulated some data: MU0 = 100 SIGMA0 = 10 MU1 = 5 NUM_USERS = 18 NUM_TRAIN_USERS = 10 MEAN_NUM_OBS = 5 def expand_list_of_list(lol): return [x for y in lol for x in y] user_mea…

Have you tried the suggestion here of also resetting the dimension of the model [image] "shape mismatch" when new data is set as a predictor for sample_posterior_predictive v5 With a simple linear regression, pymc.sample_posterior_predictive returns error: shape mismatc…

So I have not been able to get this to work, but I believe this is on the right track. The issue is that you are trying to make inferences on groups outside of the model. In this case, I think you need to actually define a second model and then call sample_posterior_predictive to make predictions on…

Thanks for the suggestion @iavicenna . In the original code I have already set the dims in the observed variable. But I guess maybe I need to also do it for other variables? Will look at this some more.

@JAB thanks for the detailed response! Let me try your suggestions and I will report back.

How to make inference on new data with a hierarchical model?

Questions v5

iavicenna May 3, 2024, 11:46am 2

Have you tried the suggestion here of also resetting the dimension of the model

Topic		Replies	Views
Predict on unseen group in hierarchical model Questions	6	1618	November 9, 2020
How to properly do out-of-sample prediction for hierarchical model v5 modeling , hierarchical , prediction	24	1653	January 25, 2024
Out of sample/model predictions in hierarchical model with new observed data v5 modeling	9	527	October 16, 2024
How do we predict on new unseen groups in a hierarchical model in PyMC3? Questions	10	5505	September 12, 2022
Out of sample prediction with new category version agnostic	5	996	August 3, 2022

How to make inference on new data with a hierarchical model?

Related topics