BMNExamples: Sampling data using Gibbs sampling for use in the examples
In BMN: The pseudo-likelihood method for pairwise binary markov networks

Description Usage Arguments Details Value Author(s) See Also

Sampling from pairwise binary Markov model using Gibbs sampling. This function is not efficient and only intended to be used in the examples.

1	BMNSamples(Theta, numSamples, burnIn, skip)

`Theta`	Parameter matrix for the model from which the data is being generated.
`numSamples`	Number of samples to return.
`burnIn`	Number of samples to discard as burn in.
`skip`	Number of samples to discard in-between returned samples.

BMNSamples generates numSamples by using Gibbs sampling. When using Gibbs sampling, it is necessary to discard the initial samples, which is controlled by the parameter burnIn. In order for the drawn samples to be independent, samples in-between also have to be discarded, which is controlled by skip.

Returns a matrix of 0 and 1 of size numSamples times p where p is the number of rows of Theta.

Holger Hoefling

BMNPseudo, BMNExact

BMN documentation built on May 29, 2017, 10:55 a.m.