mmsscore: Calculate leave-n-out cross validation (LNOCV) score
In reumandc/mms: Model Selection Via Resampling for Matrix Models

Description Usage Arguments Value Author(s) Examples

This function is used to calculate an LNOCV score for a given model, based on mean squared error, out of sample

1	mmsscore(mats, pred, n, maxruns)

`mats`	A list of numeric matrices, all assumed to be the same dimensions and symmetric. Diagonals are not used. Off-diagonal non-finite values not allowed. The first entry taken to be the response.
`pred`	The indices in mats of predictor variables in the more complex of the two models to be compared, should not include 1. Input is numeric value(s), e.g. pred=2, pred =2:3, pred =c(2,3,5).
`n`	The number of sampling locations to leave out, must be at least 2, not more than `dim(mats[[1]])[1]`.
`maxruns`	The maximum number of leave-n-outs (LNOs) to do. To be used if choose(dim(mats[[1]]),n) is very large. Inf to use (or try to use) all LNOs. If maxruns is a number, then LNOs are selected randomly and hence may include repeats.

mmsscore return an object of class list consisting of

`lno.score`	The out-of-sample forecast accuracy (mean squared error)
`num.pos`	The possible number of LNOs for the given n and number of locations
`num.att`	The total number of LNOs attempted
`num.rnk`	The number of LNOs that did not result in a rank deficiency regression problem, and so could be used for testing out-of-sample predictions
`num.usd`	The number of LNOs that could be used in the end (possibly less than num.rnk because of NAs in the input matrices)

Tom Anderson, anderstl@gmail.edu; Daniel Reuman, reuman@ku.edu; Jon Walter, jaw3es@virginia.edu

v2<-matrix(rnorm(100),10,10)
v2<-v2+t(v2)
v3<-matrix(rnorm(100),10,10)
v3<-v3+t(v3)
v4<-matrix(rnorm(100),10,10)
v4<-v4+t(v4)
err<-matrix(rnorm(100,sd=.1),10,10)
err<-err+t(err)
v1<-1*v2+2*v3+3*v4+1+err
mats<-list(v1=v1,v2=v2,v3=v3,v4=v4)
pred<-2:4
n<-2
maxruns<-Inf
h<-mmsscore(mats=mats,pred=pred,n=n,maxruns=maxruns)