Consider the model Y: Observed variables following a multivariate Normal distribution with shape (N, K) i.e. N observations with dimension K. mu: A vector representing the mean of the multivariate Normal, with shape (K,). cov: A matrix representing the covariance of the multivariate Normal, with sh…

This doesn’t quite fit with my problem unfortunately. The unique issue is that for each observation vector, some of the values may be missing. Hence for these observations, the likelihood only depends on a submatrix of the covariance and a sub vector of the mean. I imagine I could have a for loop …

[image] Woger_W: The unique issue is that for each observation vector, some of the values may be missing. If they are all unique, then yes you need to use something like a for-loop to build the logp for each vector - the solution above is for when you have multiple slices with the same imputa…

Dealing with partially observed multivariate Normal values

Questions

junpenglao January 5, 2019, 3:13am 2

You can have a look at the solution in How to marginalize on a Multivariate Normal distribution. Also see a related discussion here: Multivariate Normal with missing inputs

Topic		Replies	Views
What does the hierarchical model look like when having missing in observed? Questions	12	1197	October 31, 2018
Handling missing values in predictor when outcome is a Multivariate Normal distribution v5	7	207	October 25, 2024
Multivariate Normal with missing inputs Questions	9	1605	December 27, 2018
Multivariate normal with missing data imputation operands could not be broadcast together with shapes (29,2) () (29,) Questions theano	12	1947	September 7, 2020
How to marginalize on a Multivariate Normal distribution Questions	13	8263	November 23, 2018

Dealing with partially observed multivariate Normal values

Related topics