sim.n: Simulate n sightings from NHPP
In david-borchers/LT2D: Line transect estimation in two dimensions

Description Usage Arguments Details Value Examples

Simulates n sighting locations (x,y) given a perp.dist distribution model and detection hazard model

1 2	sim.n(n, ymin, ystart, w, hr, b, pi.x, logphi, fix.n = TRUE, intscale = NULL, nbuffer = NULL)

`n`	sample size
`ymin`	smallest forward distance
`b`	vector of detection hazard function parameters
`pi.x`	perpendicular distance distribution function
`logphi`	vector with log of pi.x parameters
`fix.n`	if TRUE sample size of exactly n is generated, else sample is generated from model with expected sample size n
`intscale`	amount by which to multiply detection location pdf f(x,y) in order to get required sample size. Either an object of class "ppscale" output by `calc.lpars` or NULL (in which case `calc.lpars` is called inside `sim.n`.
`nbuffer`	amount by which to multiply the expected n by (given all model parameters and intscale) to reduce probability that generated n is less that n on first call to NHPP generating funciton rpoispp. If NULL, it is set to 1.25 inside `sim.n`.
`ymax`	largest forward distance
`W`	perpendicular truncation distance
`hfun`	detection hazard function

Uses the spatstat function rpoispp to generate detections from a NHPP, with intensity parameter such that the expected (if fix.n is FALSE) or actual (if fix.n is TRUE) sample size is n.

a list object with element 1 a data.frame of simulated x and y sightings locations; element 2 spatstat object of class "ppp" with x- and y-coordinates of detections and element 2 simulation settings, comprising of n

## Not run: 
# simulate with fixed n:
n=100;ymin=0.01;ymax=5;W=2
b=log(c(0.75,1));logphi=c(0.5,log(0.3))
dat=sim.n(n,ymin,ymax,W,h2,b,pi.norm,logphi)
dat$locs$n

plot(density(dat))
contour(density(dat),add=TRUE)
plot(dat,pch="+",cex=0.75,add=TRUE)
hist(abs(dat$y),xlab="Perpendicular distance",main="")
hist(dat$x,xlab="Forward distance",main="")
# do same with random n:
dat=sim.n(n,ymin,ymax,W,h2,b,pi.norm,logphi,fix.n=FALSE)

# compare time taken if calculate intscale on the run vs pass it:
# first calculate each time:
system.time(for(i in 1:20) dat<-sim.n(n,ymin,ymax,W,h2,b,pi.norm,logphi))
# then calculate once and pass:
intscale=calc.lpars(n,ymin,ymax,W,h2,b,pi.norm,logphi)
system.time(for(i in 1:20) dat<-sim.n(n,ymin,ymax,W,h2,b,pi.norm,logphi,intscale=intscale))

## End(Not run)