tests/1-make-Q.R

In antiphon/PenGE: Penalised Gibbs Estimation

# Test make-Q
devtools::load_all(".")

library(spatstat)
data(lansing)

x <- lansing
library(PenGE)
if(1){
p <- 6 # we have six species
rv <- c(0.05, 0.1) # use this for everything
r1 <- rep( list(rv) , p) # intra-type interaction ranges
r2 <- rep( list(rv), p * (p-1)/2) # inter-type
Q <- make_Q_stepper_multi(x=x, ranges1 <-r1 , ranges2 <- r2, verb = 1)
# fit
fit <-fitGlbin_CV(Q, verb = 1)
}

M <- imatrix(fit, signed = TRUE)
image.imatrix(M, col = c("red","black","green"))