I have multivariate censored (clipped) data, which I am currently modeling using y = pm.MvNormal(name="$Y$", mu=mu, cov=Sigma, observed=items, shape=items.shape) From this example it seems like one simply has to add two terms to the log likelihood. Number of samples at the lower bound multiplied…

IIUC there is no close form expression of MvN CDF: https://en.wikipedia.org/wiki/Multivariate_normal_distribution#Cumulative_distribution_function

Multivariate model with censored data

junpenglao May 14, 2020, 5:30am 4

High dimension stuff is difficult

1 Like

Topic		Replies	Views
Dealing with imputation of left censored multivariate normal distribution Questions	7	1637	June 20, 2018
Modelling censored data in pymc v4: a simple Gaussian random effects example v5 development , modeling	0	398	August 30, 2022
Multivariate normal CDF v5	2	403	October 8, 2022
Call MvNormal.random() method inside model Questions	3	1159	August 29, 2018
Dealing with partially observed multivariate Normal values Questions	3	1685	January 7, 2019