getBestModelIndices: getBestModelIndices

Description Usage Arguments Author(s) See Also Examples

Description

select the best models based on (temperated) logDensity components

Usage

1
getBestModelIndices(resLogDen, dInfos, prob = 0.1, doOrder = FALSE)

Arguments

resLogDen

numeric matrix (nStep x nResComp): logDensity (highest are best)

dInfos

list of lists with entry resCompPos (integer vector) specifying the position of result components for each density

prob

proportion of good indices to return

doOrder

if set to TRUE results will be ordered, yield a slight performance penalty

Author(s)

Thomas Wutzler

See Also

orderLogDen

Examples

1
2
3
4
logDenT <- cbind( -sample(5)/2, -sample(5), -sample(5) )
#dInfos <- list( d1=list(resCompPos=1:2), d2=list(resCompPos=3) )
dInfos <- list( d1=list(resCompPos=2), d2=list(resCompPos=3) )
getBestModelIndices(logDenT, dInfos, 0.2)

twDEMC documentation built on May 2, 2019, 5:38 p.m.