Hi all, I’m trying to build a multilevel model that should estimate the electricity consumption of a building based on: an intercept that depends on the day of the week, a temperature slope coefficient for days when cooling is needed, a temperature slope coefficient for days when heating is needed…

Hi again, sorry for the delayed response! Here’s what I would try, if I understand you correctly: with pm.Model(coords=coords) as varying_intercept_and_temp: cluster_idx = pm.Data("cluster_idx", cluster, dims="obs_id") heating_temp = pm.Data("heating_temp", temp_dep_h, dims="obs_id") c…

Multilevel model with covariance of 3 variables

Questions

Sayam753 January 16, 2021, 2:06pm 7

Yes. The same issue was reported in #4391 and fixed by @Spaak . PyMC3 3.11.0 release will include this fix

2 Likes

Topic		Replies	Views
Correlated slopes in multivariate model Questions	9	3478	July 14, 2018
How to build a multi-covariance model in pymc3? Questions	2	1232	October 7, 2020
Random Intercepts and Slopes (Correlated) Convergence Issues Questions	16	2063	March 11, 2019
Multi-level multi-variables model Questions	4	1240	January 5, 2018
Specify covariance between intercepts and slopes in a 'no pooling' model Questions	6	621	December 16, 2020

Multilevel model with covariance of 3 variables

Related topics