sample_size_sim: Simulating the effect of sample size on inference with...

In robertness/bninfo: Queries and information theoritic operations on Bayesian networks.

Description

Simulating the effect of sample size on inference with constraint based algorithms. Given a parameterized Bayesian network, this simulation simulates data of a given sample size, learns undirected edges from the simulated data using a constraint-based algorithm, and collects the number of tests used in the inference as well as the number of true and false positives. Several iterations are conducted and the median number of tests, fp count and tp count is returned.

Usage

sample_size_sim(net, N, algo, alpha = 0.05, m = 100, verbose = FALSE)

Arguments

net	a parameterized Bayesian network of class bn.fit
m	the number of iteration
a	numeric vector of sample size values
a	constraint-based algorithm gs, iamb, fast.iamb, inter.iamb
the	target nominal type I error rate

robertness/bninfo documentation built on May 27, 2019, 10:32 a.m.