probit.fct: Mrds probit detection and related functions

In hierarchicalDS: Functions to Perform Hierarchical Analysis of Distance Sampling Data

Description

For independent observers, probit.fct computes observer-specific detection functions, conditional detection functions, delta dependence function, duplicate detection function (seen by both), and pooled detection function (seen by at least one).

Usage

probit.fct(x, formula, beta, rho, ...)

Arguments

x	vector of perpendicular distances
formula	linear probit formula for detection using distance and other covariates
beta	parameter values
rho	maximum correlation at largest distance
...	any number of named vectors of covariates used in the formula

Details

The vectors of covariate values can be of different lengths because expand.grid is used to create a dataframe of all unique combinations of the distances and covariate values and the detection and related values are computed for each combination. The covariate vector observer=1:2 is automatically included. The folowing is too long for the examples section: test=probit.fct(0:10,~distance,c(1,-.15),.8,size=1:3) par(mfrow=c(1,2)) with(test[test$observer==1,], plot(distance,p,ylim=c(0,1),xlab="Distance",ylab="Detection probability") points(distance,pc,pch=2) points(distance,dup,pch=3) points(distance,pool,pch=4) legend(1,.2,legend=c("Detection","Conditional detection","Duplicate detection","Pooled detection"),pch=1:4,bty="n") plot(distance,delta,xlab="Distance",ylab="Dependence") )

Value

dat dataframe with distance, observer, any covariates specified in ... and detection probability p, conditional detection probability pc, dupiicate detection dup, pooled detection pool and dependence pc/p=delta.

Author(s)

Jeff Laake

hierarchicalDS documentation built on July 3, 2019, 1:07 a.m.