AS.test {abctools} | R Documentation |
The function tests to determine adding a (set of) statistics is informative in ABC inference.
AS.test(grid = 10, x1, x2, supp=NULL)
grid |
the number of bins into which to divide the posterior sample for the approximate sufficiency calculation. |
x1 |
the posterior sample using the first set of summary statistics. |
x2 |
the posterior sample using the second (alternative) set of summary statistics. |
supp |
the "support" of the prior (e.g. uniform bounds). |
After dividing each posterior sample into a number of bins (specified by grid
), the function computes the ratio of the posterior densities. This is seen as a measure of information added (sufficiency) by using the alternative posterior sample instead of the first posterior sample. If the ratio exceeds a particular threshold (a number of standard deviations away from the expected counts in each bin), then the alternative set of summaries is seen as being more informative.
extreme |
a boolean value indicating whether the alternative posterior sample is more informative than the first (i.e. the extra summary statistics add information. |
Matt Nunes
Blum, M. G. B, Nunes, M. A., Prangle, D. and Sisson, S. A. (2013) A
comparative review of dimension reduction methods in approximate
Bayesian computation. Stat. Sci. 28, Issue 2, 189–208.
Joyce, P. and P. Marjoram (2008) Approximately sufficient statistics and
Bayesian computation. Stat. Appl. Gen. Mol. Biol. 7
Article 26.
Nunes, M. A. and Prangle, D. (2016) abctools: an R package for tuning
approximate Bayesian computation analyses. The R Journal
7, Issue 2, 189–205.
#create two fake posterior samples: x1<-runif(10000) x2<-rnorm(10000) AS.test(x1=x1,x2=x2,supp=range(x2))