CODEWITHSUNDEEP
rspatstat
jbeds
jbeds

Reputation: 17

Trying to simulate point patterns via simulate.mppm causes error

I am trying to simulate a point pattern based on a model fit to a dataset of several multitype point patterns. However, I can't run simulate.mppm without raising an error. I attached the code below (using optimbase_1.0-9 and spatstat_1.62-2):

library(spatstat)
library(optimbase)
flusubset <- flu[1:4]
typelist <- lapply(lapply(flusubset$pattern, marks), levels)
num_marks <- length(typelist[[1]])
iradii <- 50*ones(num_marks,num_marks)
hradii <- 3*ones(num_marks,num_marks)
Int <- anylist()
for (i in 1:dim(flusubset)[1]) {
  Int[[i]] <- MultiStraussHard(iradii=iradii, hradii=hradii)
}
Int <- as.hyperframe(Int)
multmodel <- mppm(pattern ~ 1, data=flusubset, interaction=Int)
simulate(multmodel)

Whenever I run this code, I encounter the error:

Error in check.nvector(w, nrow(x), things = "rows of x") : The length of ‘w’ (=16744) should equal the number of rows of x (=15424)
9. stop(whinge)
8. check.nvector(w, nrow(x), things = "rows of x")
7. sumouter(mom, lam * wt * glmsub)
6. vcmGibbs(object, ..., what = what, err = err)
5. vcov.mppm(object, what = "fisher", err = "null")
4. vcov(object, what = "fisher", err = "null")
3. subfits(object)
2. simulate.mppm(multmodel)
1. simulate(multmodel)

Thank you for the help!

Upvotes: 0

Views: 42

Answers (1)

Adrian Baddeley
Adrian Baddeley

Reputation: 2973

This is a bug in current versions of spatstat.

I have fixed it in the development version spatstat 1.62-2.011 available from the GitHub repository. The next public release of spatstat containing the bug fix will be in February 2020.

For bugs in spatstat, please post an Issue on the issues page at the GitHub repository.

Upvotes: 1

Related Questions