Hi, I’m having some issues understanding how to add new unseen data into my model, to see how it would effect my posterior. I’ve looked at these posts ( here , here but I still have not been able to understand what I’m doing wrong. Hopefully this explanation will be enough to understand what the is…

It looks like you are confusing the posterior predictive with the posterior. val are the observations, so changing them has no effect on posterior predictive sampling. sample_posterior_predictive uses the samples from the posterior p(\theta|y) to sample from the posterior predictive p(\tilde{y}|y) =…

Using pm.Data to predict on two inputs for sample_posterior_predictive; why is there no change in the results?

Questions

chef1075 May 17, 2021, 10:55am 5

Ah this is quite helpful. Thanks @ricardoV94 !

Topic		Replies	Views
How do I predict on new, unseen real data using pm.sample_posterior_predictive? Questions	13	8589	January 7, 2021
Simulating new data points while having data-dependent priors Questions	0	453	May 10, 2019
Sample_posterior_predicitve not catching shape of new data v5 prediction	10	1321	August 24, 2022
11.16 Rethinking Code Questions	10	948	June 4, 2020
Automatic imputation and posterior predictive sampling with new data Questions	5	2068	October 29, 2020

Using pm.Data to predict on two inputs for sample_posterior_predictive; why is there no change in the results?

Related topics