Closed YimenAraya-Ajoy closed 2 years ago
This is a great idea, and I have thought about this a lot. I think it becomes prohibitively difficult when we allow people to specify their own formulas. To do this I think we have to be able to anticipate what kinds of model formulations people are going to be able to simulate. We could implement this quite easily when a formula isn't specified (things are just added up and assumed to be uncorrelated), and definately in the squidApp implementation
implemented in b7ba826 with the new expected_variance() function. Is this what you had in mind @YimenAraya-Ajoy ?
need to make it work with multivariate
I think it will be nice to have a function that summarizes the expected patterns of variance and covariance at the different levels due to known and unknown sources. As well as the proportion of (co)variance explained by each process. This is something that is implemented in the squidApp