I am trying to create an observed variable in PyMC3 which has a multivariate normal distribution. I want the covariance matrix whose elements are other random variables. As an example, consider the following code: import pymc3 as pm with pm.Model() as model: a = pm.Normal('a', mu=0, sigma=10) …

You can use import theano.tensor as tt tt.concatenate([a, b, b, c]).reshape((2, 2)) Or alternatively tt.stack. If you want priors on covariance matrices, you probably want to look into pm.LKJCholeskyCov instead.

Construct matrix with random variables as elements

Questions

sojohan April 13, 2020, 1:48pm 5

Thanks

Topic		Replies	Views
Cannot convert <class 'numpy.ndarray'> to TensorType' Questions	5	4021	February 13, 2020
Model multivariate normal with separate means, dimension mismatch error Questions	2	3090	August 7, 2017
Adding two covariance matrices of different sizes Questions theano	8	2265	July 2, 2017
Draw number of MvNormal in once Questions	2	575	April 5, 2019
How to build a multi-covariance model in pymc3? Questions	2	1232	October 7, 2020

Construct matrix with random variables as elements

Related topics