psiLogOdds: psiLogOdds

In dlandy/WarpedBayes: Implements Design Pattern Discussed in Landy, Crawford, & Corbin

Description

This function is a transformation function that takes the log odds By design, it is robust to the inclusion of the occasional 0, which it maps to a very small value, which can optionally be specified It's appropriate for mapping (0, 1):-> (-inf, inf).

Usage

psiLogOdds(stimuli, smallValue = 10^-30)

Arguments

stimuli	a vector of positive stimuli, between 0 and inf
smallValue	A value to set nominal 0's to, to avoid errors in plotting simulated data, or aberrant responses

Value

a vector containing "warped" stimuli

See Also

psiLinear, psiLogOdds, psiPrelec, psiLog, psiIdentityInverse

Examples

psiLogOdds(1:100/100)
(0:1000/1000) %>% psiLogOdds() %>%
     vanillaBayes() %>% 
     psiLogOddsInverse()  # Implements Gonzales & Wu, 1996

dlandy/WarpedBayes documentation built on May 29, 2019, 2:49 p.m.