predictiveStats: Predictive Analysis for a fitmat3 object

In guiludwig/mat3c: Marked spatial point processes with inhibition based on Matern-III clustered marks

Description

Evaluates the predictive posterior probability of T > T_obs for various statistics.

Usage

predictiveStats(model, fname = NULL, whichSamples,
  additionalArgs = list(win = owin(), R = 0.005, R_c = 0.02), ...)

Arguments

model	An fitmat3 object (resulting from a call to fitmat3)
fname	If the MCMC samples were saved into text files, this can be used to generate predictive probabilities from them. Defaults to NULL.
whichSamples	Which (and how many) samples to collect in order to generate the predictive probabilities. Must be smaller than the number of MCMC samples.
additionalArgs	additional arguments to be passed to rmat3() (if read from text files, it is necessary to provide a list with window, R and R_c).

Value

predictiveStats returns a list containing at least the following components

pL	predictive probability for the expected number of centers
pRc	predictive probability for the number of fingers within R_centers distance of other centers
pnx	predictive probability for the average number of fingers
psx	predictive probability for the standard deviation of the number of fingers
chx	predictive probability for the mean area of the convex hull of fingers (reactive territory)
spx	predictive probability for the spanned angle of fingers
mdx	predictive probability for the average of minimum distance between fingers
adx	predictive probability for the average of average distance between fingers
cex	predictive probability for the average distance between centers

guiludwig/mat3c documentation built on Dec. 2, 2019, 1:32 a.m.