2018
DOI: 10.1111/2041-210x.13014
|View full text |Cite
|
Sign up to set email alerts
|

R2ucare: An r package to perform goodness‐of‐fit tests for capture–recapture models

Abstract: Assessing the quality of fit of a statistical model to data is a necessary step for conducting safe inference. We introduce R2ucare, an r package to perform goodness‐of‐fit tests for open single‐ and multi‐state capture–recapture models. R2ucare also has various functions to manipulate capture–recapture data. We remind the basics and provide guidelines to navigate towards testing the fit of capture–recapture models. We demonstrate the functionality of R2ucare through its application to real data.

Help me understand this report
View preprint versions

Search citation statements

Order By: Relevance

Paper Sections

Select...
3
1
1

Citation Types

0
52
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
8

Relationship

0
8

Authors

Journals

citations
Cited by 91 publications
(52 citation statements)
references
References 26 publications
0
52
0
Order By: Relevance
“…Additional research is needed to test these hypotheses. Our findings emphasize the importance of checking for violations of the assumption of equal detectability in multispecies mark‐recapture studies, even when examining species thought to have highly comparable ecologies (Gimenez et al ).…”
Section: Discussionmentioning
confidence: 67%
See 2 more Smart Citations
“…Additional research is needed to test these hypotheses. Our findings emphasize the importance of checking for violations of the assumption of equal detectability in multispecies mark‐recapture studies, even when examining species thought to have highly comparable ecologies (Gimenez et al ).…”
Section: Discussionmentioning
confidence: 67%
“…In such models, Φ represents the apparent survival probability because the complementary estimate is a combination of mortality and permanent emigration. To our knowledge, no tests yet exist to assess goodness‐of‐fit (GOF) of CJS models with individual and temporal covariates (Gimenez et al ). Therefore, we summarized capture histories by frequency for each species to assess GOF for a fully time‐dependent, species‐specific CJS model.…”
Section: Methodsmentioning
confidence: 99%
See 1 more Smart Citation
“…Model convergence was confirmed both visually and by using the “R hat” Gelman–Rubin statistic (Gelman & Rubin, ). We assessed the goodness of fit of all our submodels using postpredictive checks and the R package R2ucare (Gimenez, Lebreton, Choquet, & Pradel, ) and found no clear evidence for lack of fit (Supporting Information Appendix ).…”
Section: Methodsmentioning
confidence: 99%
“…We fit Cormack-Jolly-Seber (CJS) models within program MARK to estimate the probabilities of apparent annual survival (Φ, probability of remaining alive and remaining within the study area) and recapture (ρ) from our live-encounter banding records; as is standard for CJS models, adult thrushes newly captured in 2015 (the last primary period) were not included in our modeling effort (Cormack 1964, Jolly 1965, Seber 1965, White and Burnham 1999. Goodness-of-fit tests, including for overdispersion, conducted with the R package R2ucare (Gimenez et al 2018, R Core Team 2019), suggested that our data were appropriate for CJS models (Pradel et al 2005). We considered total net hours (a measure of netting effort) across June and July (2002-2015) as our only covariate modeling recapture probability.…”
Section: Modeling Survival Hypotheses and Choice Of Covariatesmentioning
confidence: 99%