I have been working on modeling the below State Space Model X_{i,t+1} \sim N(\mu_t, Q)\\ \mu_{i,t} = Ax_{i,t} + BU_{i}\\ y_{i,t} \sim Poisson(\lambda_{i,t}) \\\lambda_{i,t} = C \exp(X_{i,t}) where X_{i,t+1} is the hidden state for ith sample at time t, U_i is known exogenous variables, \lambda_{i…

Do these Utest happen after the U in your original model? In that case you can create a new model which starts where U “left off” and call posterior predictive there. I have an example in this WIP notebook with a single timeseries, but the same concept applies. You can sample new unobserved variabl…

Posterior predictive on testing data in State Space Models

Questions v5

dushyant May 12, 2023, 12:06am 3

Thanks! This is really helpful!

1 Like

Topic		Replies	Views
Forecasting with pymc3 Questions	1	2571	November 24, 2017
Learning Poisson State Space Model parameters for large number of samples version agnostic modeling	14	720	May 10, 2023
Hidden Markov Model - Predicting the next observation(s) of unseen sequences version agnostic	8	2018	May 31, 2022
Held-out prediction with latent variables whose dimension changes with the test set Questions	1	555	April 11, 2019
Indexing of Posterior Predictions in State Space Model	2	176	February 23, 2024

Posterior predictive on testing data in State Space Models

Related topics