plotDgam: SECR Density spline plotting.

In david-borchers/secrgam: Spatially Explicit Capture-Recapture with Generalised Additive Density Models

Description

Plots regression spline smooth components of an secr density model.

Usage

plotDgam(fit, scale = "response", mask.rug = FALSE, det.rug = TRUE,
  bounds = TRUE, npts = 200, main = TRUE)

Arguments

fit	object returned by secrgam.fit.
scale	either "link" (for plot on link funtion scale) or "response" (for plot on response scale).
mask.rug	TRUE if a rugplot of locations of mask points in the covariate dimension is to be plotted.
det.rug	TRUE if a rugplot of locations of detectors in the covariate dimension is to be plotted.
bounds	TRUE if 95% confidence bounds of the smooth are to be plotted.
npts	number of points on x-axis at which to evaluate smooth.
main	if TRUE, a main heading reflecting the smooth term in the model that is being plotted, is added to the plot; else no main heading is plotted.

Details

Plots smooth on link or response scale, together with rug plot showing locations of detectors in covariate space (if det.rug==TRUE), rug plot showing locations of mask points (if mask.rug==TRUE) and 95% confidence bounds (if bounds=TRUE).

This function is not currently able to deal with interactions between sessions and other covariates. It can plot a smooth of session or of mask covariates but in the case of mult-session fits, it will plot only smooths of non-session covariates for the first session, using the firs session mask.

Be aware that using mask.rug=TRUE will result in it taking quite a while to produce the plot

Examples

op = par(no.readonly = TRUE)

data(Boland.fits1)
plotDgam(fit1.a3)

par(mfrow = c(1,2))
data(Boland.fits2)
plotDgam(fit2.a3.dW3)

par(op)

david-borchers/secrgam documentation built on May 14, 2019, 9:30 p.m.