Displaying 1 result from an estimated 1 matches for "tlambda".
Did you mean:
lambda
2011 Jan 05
0
Nnet and AIC: selection of a parsimonious parameterisation
...ize+1; j=1;
choice <- numeric(maxSize); nparam <- numeric(maxSize); lambdaj <-
numeric(maxSize)
AIC <- numeric(maxSize); nparamInit <- numeric(maxSize)
ri <- sample(nifold, nrow(data), replace = TRUE)
size=seq(maxSize-1,0); minAIC=1000000
for(j in 1:maxSize) {
tlambda=optimize(CVnn1,decayRange,tol = 0.0001,formula, data, nreps=nreps,
ri=ri, size=size[j], linout=linout, skip=skip, maxit=maxit,
optimFlag=TRUE, alpha=.5)
lambdaj[j]=tlambda[[1]]
cat(" size =", size[j], "decay =", round(tlambda[[1]],3), "\n")...