ind.masses & group.masses - TExPo & TInPo

[GoogleCodeExporter]

ind.masses are required for proper LOO computations. 

group.masses presume a baseline or a priori if they do not change. Also, in 
general, ind.masses should make a return.

This requires updates to multiple functions include supplementaryProjections 
(and related functions).

group.masses should be computed from ind.masses if ind.masses are given, and 
group.masses are not. else, group.masses are presumed to be a static 
(Bayesian-like) prior.

Original issue reported on code.google.com by jenny.ri...@gmail.com on 16 Jun 2013 at 8:48

[GoogleCodeExporter]

This was actually submitted by Derek (who stole JRs computer).

Original comment by jenny.ri...@gmail.com on 16 Jun 2013 at 8:48

[GoogleCodeExporter]

For now, ind.masses have been removed. This bug will stay open. 
As noted, there is a collision between group.masses and ind.masses if they do 
not match up. So, ind.masses will make a return later.

If ind.masses are given, they supersede group.masses and group.masses are 
computed from ind.masses.

Original comment by DerekBea...@gmail.com on 6 Dec 2013 at 5:47

• Changed state: Accepted
• Added labels: Priority-High
• Removed labels: Priority-Medium

[GoogleCodeExporter]

Actually, it is possible to put ind.masses and group.masses together but... it 
makes little sense, e.g., 

Mg %*% (Y' * (Mi %*% X)) %*% W

wherein Mg are group masses and Mi are individual masses. 

But this is so highly specific that I don't want to facilitate this via 
function -- a user can do this on their own.

Original comment by DerekBea...@gmail.com on 6 Dec 2013 at 5:51