Hi, I am new to pymc and I am trying to understand how to build time series models. I am starting with an AR1 model where I am trying to replicate what the already available AR distribution in pymc gives us. This is to help me understand how to use scan to iterate through a time series and how I am…

A few specific comments about your 3rd implementation: The concatenate fails because you need to add a dummy batch index to init. pt.concatenate([init[None], upd], axis=0]) should work. What you’re implementing is quite close to the OLS formulation of AR, which would look something like this: wit…

Implementing a manual AR1 model

Questions v5

AlexAndorra May 7, 2025, 3:46pm 4

FYI, the Bayesian Gods made @jessegrabowski’s wishes a reality, as I think this notebook is what he was calling for

1 Like

Topic		Replies	Views
Vectorized autoregressive model Questions	5	2347	January 2, 2020
Autocorrelation in a model Questions	8	1737	May 10, 2018
Applying timeseries over periodic data Questions	8	1696	November 7, 2017
Reproducing PyMC3 AR(1) tutorial in PyMC4 Development	1	983	June 7, 2020
AR model "too many indices for array" Questions	1	1784	January 18, 2019

Implementing a manual AR1 model

Related topics