You can try to use a Simulator with sample_smc, it’s the dirtiest way to do inference without a closed form likelihood, when you have a generating process.
Might not scale very well though: pymc.Simulator — PyMC dev documentation
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Looking for any resources about modeling/predicting revenue/sales | 0 | 496 | December 5, 2017 | |
| Problems with Hierachical Bayesian Model - Advertising | 16 | 1081 | August 10, 2018 | |
| Modelling a mixture of binomial and gamma | 2 | 982 | February 12, 2019 | |
| Simulating Fake Data from pymc3 model | 1 | 1783 | October 28, 2017 | |
| Possible Performance Issue Using Sparse Matrices in Bipartite Noisy-Or Model | 10 | 2631 | September 15, 2017 |