npr: Calculate a non-parametric regression surface

In ctmm-initiative/ctmm: Continuous-Time Movement Modeling

Calculate a non-parametric regression surface

Description

This function estimates the mean value of an annotated covariate as a function of location, using non-parametric regression.

Usage

npr(data,UD,variable="speed",normalize=FALSE,debias=TRUE,error=0.001,...)

Arguments

data	2D timeseries telemetry data represented as a telemetry object or list of objects.
UD	A UD object from the output of akde.
variable	Variable for mean estimation. Can be a column of data.
normalize	Consider variable as providing a weighted probability distribution.
debias	Correct for oversmoothing if normalize=TRUE.
error	Target probability error.
...	Arguments passed to akde.

Value

Returns a UD object.

Author(s)

C. H. Fleming.

See Also

akde, occurrence

Examples

# Load package and data
library(ctmm)
data(buffalo)
DATA <- buffalo$Cilla

# calculate fit guess object
GUESS <- ctmm.guess(DATA,interactive=FALSE)
# in general, you should be running ctmm.select here instead of ctmm.fit
FIT <- ctmm.fit(DATA,GUESS)

# Compute akde object
UD <- akde(DATA,FIT)

# compute revisitation distribution
RD <- revisitation(DATA,UD)

# Plot data with revisitation distribution
plot(DATA,RD)

ctmm-initiative/ctmm documentation built on April 18, 2024, 9:39 a.m.