Description Usage Arguments Details Value See Also Examples
uhcdenscalc
calculates kernal density estimates to create UHC plots.
1 | uhcdenscalc(rand_sims, dat, avail, gridsize = 500)
|
rand_sims |
The characteristics associated with the simulated locations in the test data set |
dat |
The characteristics associated with observed locations in the test data set |
avail |
The characteristics associated with available locations in the test data set |
gridsize |
The size of grid for estimating the kernel density estimate (kde). Default value is 500. |
uhcdenscalc
calculates density estimates for the environmental
characteristics associated with the observed and available locations in the
test data and also those associated with the randomly chosen locations
generated by the uhcsim
or uhcsimstrat
functions.
A list of density estimates associated with the simulated locations, (densrand), the observed locations (densdat), and the available locations (densavail)
Full archive of the data and code necessary to replicate the manuscript at http://doi.org/10.13020/D6T590.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 | # Simulate training data for the non-linear example
nonlinear.train <- uhcdatasimulator(nused = 100,
navail = 10000,
betas = c(2,-1),
ntemp = 1000000,
example = "non-linear")
# Simulate test data for the non-linear example
nonlinear.test <- uhcdatasimulator(nused = 100,
navail = 10000,
betas = c(2,-1),
ntemp = 1000000,
example = "non-linear")
# Fit GLM with quadratic relationship
train.correct <- glm(y~temp + I(temp^2),
family = binomial,
data = nonlinear.train)
# Fit GLM with linear (misspecified) relationship
train.misspec <- glm(y~temp,
family = binomial,
data = nonlinear.train)
# Simulate data for quadratic model
xhat.correct <- uhcsim(nsims = 1000,
nused_test = 100,
xmat = model.matrix(y~temp + I(temp^2), data = nonlinear.test)[,-1],
fit_rsf = train.correct,
z = as.matrix(nonlinear.test[,"temp"]))
# Simulate data for linear (misspecified) model
xhat.misspec <- uhcsim(nsims = 1000,
nused_test = 100,
xmat = as.matrix(model.matrix(y~temp, data = nonlinear.test)[,2]),
fit_rsf = train.misspec,
z = as.matrix(nonlinear.test[,"temp"]))
# Get density estimates for quadratic model
denshats.correct <- uhcdenscalc(rand_sims = xhat.correct[,,1],
dat = subset(nonlinear.test, y==1, select="temp"),
avail = subset(nonlinear.test, y==0, select="temp"))
# Get density estimates for linear (misspecified) model
denshats.misspec <- uhcdenscalc(rand_sims = xhat.misspec[,,1],
dat = subset(nonlinear.test, y==1, select="temp"),
avail = subset(nonlinear.test, y==0, select="temp"))
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.