getIC: New way of finding Bin ! (Now A)

In Marie-PerrotDockes/sanssouci.hmm: Confindence bound on the FDP after post-selection under an HMM

New way of finding Bin ! (Now A)

Description

New way of finding Bin ! (Now A)

Usage

getIC(m, alpha, li0, f0x, f1x, Pis)

Arguments

m	the number of positions (hypothesis)
alpha	a matrix m * 2 containing the forward variables
f0x	a vector of the values of the density under the null hypothesis on the observations
f1x	a vector of the values of the density under the alternative hypothesis on the observations
A	a matrix 2 * 2 the transition probabilities
beta	a matrix m * 2 containing the backward variables

Value

Product of matrices

Examples

m <-  100
A <- matrix(c(0.95, 0.05, 0.2, 0.80), 2, 2, byrow = T)
  rdata <- sim_hmm_2states(m, Pi, A, f0 = c(0,1), f1= c(2,1))
 x <- rdata$x
 f0x <- dnorm(x)
 f1x <- dnorm(x, 1,2)
  mod <- for_back(m, A, f0x, f1x, Pi)
    Pis_est <- lapply(2:m, function(i){
 get_A( m,alpha = mod$alpha, beta = mod$beta, A, f0x, 
        f1x, i = i)
 })
 alpha <- 0.1
getIC(length(x), alpha, mod$gamma[,1], f0x, f1x, Pis_est)
quant <- get_quantiles(sel = 1:m, li0 = mod$gamma[, 1], 
 Pis = Pis_est, f0x = f0x, f1x = f1x)
 borne(type_borne = "HMM", sel = 1:m, a = quant, alpha )
 borne(type_borne = "HMM_small", sel = 1:m, a = quant, alpha)

Marie-PerrotDockes/sanssouci.hmm documentation built on Oct. 26, 2023, 10:36 a.m.