Hi, I do have a rather general question regarding missing values in edward/edward2. In biostatistical applications data is often missing. A standard approach is (multiple) imputation of the missing values before moving to the actual analysis. This is a bit absurd when using edward as the models use…

Thanks for the quick reply! Well, that’s possible in principle but think about a 50 dimensional longitudinal dataset with missing values in all variables and it quickly get extremely ugly to do so in practice. I just don’t see how this solution scales - or maybe I just missed the point :wink:

If your model already has local latent variables, and you can do mini-batch SGD, it should scale about as well as anything.

Sorry for being imprecise: the issue will not be scaling computationally but programmatically. There will be huge amounts of boiler-plate code if you need to do this manual splicing for more than 3 variables.

Ah, ok. Are you sure? The index variable I, in the example, is supposed to have the same shape as your training data. So if your data is N x 50, so is I - it will be zero everywhere except at the indices where data is missing, there it will be 1.

Handling missing values

deoxyribose April 12, 2018, 12:48pm 2

Maybe you can use the 2. approach Dustin describes here: How to handle missing values in Gaussian Matrix Factorization

Topic		Replies	Views
How to handle missing values in Gaussian Matrix Factorization	2	2998	April 24, 2017
Will it support Stochastic Process in future?	5	1762	December 12, 2017
Error using Gibbs sampling for inference in a probabilistic graphical model	2	1265	May 28, 2017
HMM Implementation with Marginalized Latent States	3	2206	August 25, 2018
Implementing mixed membership models	1	885	November 9, 2017

Handling missing values

Related topics