predictFBLR: Predict Case Probabilities with Full Bayesian Logic...

In scrime: Analysis of High-Dimensional Categorical Data Such as SNP Data

Description

Predicts case probabilities for binary data (usually SNP data dichotomized with snp2bin) based on an MCMC sample of Bayesian logic regression models obtained with fblr.

Usage

predictFBLR(file, bin, kmax = 10, int.level = 2)

Arguments

file	character string naming file where MCMC sample is stored.
bin	matrix of binary variables to make predictions for. One row is one observation. The number of binary variables has to be the same as used in fblr.
kmax	the maximum number of allowed logic predictors used in fblr.
int.level	the maximum number of allowed binaries in a logic predictor used in fblr.

Value

Vector of length nrow(bin) with predicted case probabilities.

Author(s)

Arno Fritsch, [e-mail]

See Also

fblr

Examples

## Not run: 
# Use fblr on some simulated SNP data
snp <- matrix(rbinom(500 * 20, 2, 0.3), ncol = 20)
bin <- snp2bin(snp)
int <- apply(bin,1,function(x) (x[1] == 1 & x[3] == 0)*1)
case.prob <- exp(-0.5+log(5)*int)/(1+exp(-0.5+log(5)*int))
y <- rbinom(nrow(snp),1,prob=case.prob)
fblr(y, bin, niter=1000, nburn=0)

# Prediction for some new observations 
newbin <- snp2bin(matrix(rbinom(100 * 20, 2, 0.3), ncol = 20))
predictFBLR("fblr_mcmc.txt",newbin)


## End(Not run)

scrime documentation built on May 2, 2019, 10:24 a.m.