overall_JMV: Overall goodness-of-fit test for the Jolly-Move model

In R2ucare: Goodness-of-Fit Tests for Capture-Recapture Models

Overall goodness-of-fit test for the Jolly-Move model

Description

This function performs the overall goodness-of-fit test for the Jolly-Move model. It is obtained as the sum of the 5 components Test3G.SR, Test3G.SM, Test3G.WBWA, TestM.ITEC, TestM.LTEC. To perform the goodness-of-fit test for the Arnason-Schwarz model, both the Arnason-Schwarz (AS) and the Jolly-Move models need to be fitted to the data (to our knowledge, only E-SURGE can fit the JMV model). Assuming the overall goodness-of-fit test for the JMV model has produced the value stat_jmv for the test statistic, get the deviance (say dev_as and dev_jmv) and number of estimated parameters (say dof_as and dof_jmv) for both the AS and JMV models. Then, finally, the p-value of the goodness-of-fit test for the AS model is obtained as 1 - pchisq(stat_as,dof_as) where stat_as = stat_jmv + (dev_as - dev_jmv) and dof_as = dof_jmv + (dof_jmv - dof_as)

Usage

overall_JMV(X, freq, rounding = 3)

Arguments

X	is a matrix of encounter histories
freq	is a vector of the number of individuals with the corresponding encounter history
rounding	is the level of rounding for outputs; default is 3

Value

This function returns a data.frame with the value of the test statistic, the degrees of freedom and the p-value of the test.

Author(s)

Olivier Gimenez <[e-mail]>, Roger Pradel, Rémi Choquet

Examples

# read in Geese dataset
library(RMark)
geese = system.file("extdata", "geese.inp", package = "R2ucare")
geese = convert.inp(geese)

geese.hist = matrix(as.numeric(unlist(strsplit(geese$ch, ''))),nrow=nrow(geese),byrow=TRUE)
geese.freq = geese$freq

# encounter histories and number of individuals with corresponding histories
X = geese.hist
freq = geese.freq

# load R2ucare package
library(R2ucare)

# perform overall gof test
overall_JMV(X, freq)

R2ucare documentation built on July 11, 2022, 9:06 a.m.