WLR_fit: Fit weighted logistic regression model to use-nonuse and...
In qureshlatif/WoodpeckerHSI: Woodpecker habitat suitability models

Description Usage Arguments Details Value Author(s) References See Also

Fits weighted logistic regression to binary data where 1 = habitat use observation and 0 = non-use observation. Weights are set so that the sum of weights for 0s will equal the number of 1s, and sample size will be twice the number of 1s. This type of model was used to develop habitat suitability models using nest location data for disturbance-associated woodpeckers (Russell et al. 2007, Latif et al. 2016).

1	WLR_fit(dat, formula, Obs = "Nest", w = NULL)

`dat`	Data frame with one column indicating which observations are use (1) versus non-use (0), and additional columns containing habitat covariates to be used as predictors in the model.
`mod`	A character string containing the right hand side of the desired model, i.e., of the form "x1+x2+...+xK", where 'xK' = the last covariate in the model.
`Obs`	The name of the column in 'dat' indicating use-nonuse, i.e., y in the formula y~x.
`w`	Optional. If provided, must be a numeric vector of length = nrow(dat), and will used as observation weights instead of the default weights described above.

The model is fit using glm() with family = "binomial" and weights specified as described above.

Model object generated by glm()

Quresh S. Latif, Rocky Mountain Research Station, U.S. Forest Service

Latif, Q. S., V. A. Saab, J. P. Hollenbeck, and J. G. Dudley. 2016. Transferability of habitat suitability models for nesting woodpeckers associated with wildfire. The Condor 118:766-790.

Russell, R. E., V. A. Saab, and J. G. Dudley. 2007. Habitat-suitability models for cavity-nesting birds in a postfire landscape. Journal of Wildlife Management 71:2600-2611.

glm, predict.glm

qureshlatif/WoodpeckerHSI documentation built on May 29, 2019, 7:51 a.m.