I’m trying to implement a correlated random walk over the slope and intercept parameters of a linear model. I’m basing it loosely off the multivariate random walk tutorial on the pymc site . However, I keep running into some errors when trying to sample the prior predictive and posterior distribution…

Well, it looks like this was a simple mistake in my priors… sigma = pm.Normal(...) obviously should have been: sigma = pm.HalfNormal(...) Thanks for the help though!

Multivariate Random Walk Regression

Questions v5

ricardoV94 December 8, 2023, 9:43am 2

You can call .eval() on that shape variable or any.model variable to get it’s concrete value

Topic		Replies	Views
Trouble with Initilization and Posterior Sampling Questions theano	13	2225	June 15, 2018
Geometric variable not being properly Sampled Questions from_github	22	3328	June 22, 2017
Call MvNormal.random() method inside model Questions	3	1159	August 29, 2018
Multiple independent random walks with differing innovations Questions	3	634	January 8, 2021
Excessive memory - Multiple regression Questions	9	2931	October 24, 2017

Multivariate Random Walk Regression

Related topics