I am trying to use a two-step procedure to estimate some ancillary variables and then use these estimated values in a real run: the procedure is as follows: with pm.Model(coords=coords) as baseModel: # input data NDVI_ = pm.MutableData("NDVI_", NDVI) T_ = pm.MutableData("T11_", T_obs)…

Thank you again for all the feedback. Based on @iavicenna idea, I just found out a solution. First, I want to apologize because I didn’t make clear from the beginning that sm is a vector. In fact is a time series, whose value is known for some period (calibration) and must be estimated in the second…

Problem using previous samples to new inference in new model context

Questions v5

ricardoV94 July 3, 2024, 6:02am 4

If you look for “prior from posterior” in this forum you’ll find a couple of threads on the topic

Topic		Replies	Views
How to get "on-line" prediction estimates for data over time? Questions	6	803	May 19, 2020
Feed a posterior as an input random variable in another model Questions	1	342	October 6, 2020
Using pm.Data to predict on two inputs for sample_posterior_predictive; why is there no change in the results? Questions	4	1183	May 17, 2021
Problem in post predictive sampling in context of updating priors for sequential data Questions	3	507	December 21, 2018
How to predict new values on hold-out data Questions	24	13494	July 22, 2020

Problem using previous samples to new inference in new model context

Related topics