BennMacdonald / AGM_RPackage

1 stars 0 forks source link

Needs to be able to deal with unobserved variables #7

Closed FrankD closed 7 years ago

FrankD commented 7 years ago

I will try implementing this as a latent Gaussian process.

FrankD commented 7 years ago

I've implemented it; it doesn't throw any errors and seems to infer sensible parameters on the V&G model when I leave the first species out, but we should do a wider range of tests to make sure this works as intended. I will close this issue and add that to the Todo list.