IsingPL

Computes the pseudolikelihood of a dataset given an Ising Model.

Sample states from the Ising model and compute the probability of states. Sampling can be done for any number of nodes, but due to the intractibility of the Ising model the distribution can only be computed up to ~10 nodes.

Sacha Epskamp

IsingSampler

Sampling Methods and Distribution Functions for the Ising Model

Jesse Boot

IsingPL function

<dl> <dt>x</dt>
<dd>A binary dataset</dd> <dt>graph</dt>
<dd>Square matrix indicating the weights of the network. Must be symmetrical with 0 as diagonal.</dd> <dt>thresholds</dt>
<dd>Vector indicating the thresholds, also known as the external field.</dd> <dt>beta</dt>
<dd>Scalar indicating the inverse temperature.</dd> <dt>responses</dt>
<dd>Response options. Typically set to <code>c(-1L, 1L)</code> or <code>c(0L, 1L)</code> (default). Must be integers!</dd></dl>

Arguments

Sacha Epskamp (mail@sachaepskamp.com)

Author

Pseudo-likelihood — IsingPL

<dl>

 <dt>x</dt>
<dd>A binary dataset</dd>

 <dt>graph</dt>
<dd>Square matrix indicating the weights of the network. Must be symmetrical with 0 as diagonal.</dd>

 <dt>thresholds</dt>
<dd>Vector indicating the thresholds, also known as the external field.</dd>

 <dt>beta</dt>
<dd>Scalar indicating the inverse temperature.</dd>

 <dt>responses</dt>
<dd>Response options. Typically set to <code>c(-1L, 1L)</code> or <code>c(0L, 1L)</code> (default). Must be integers!</dd>

</dl>

IsingPL: Pseudo-likelihood

Description

Usage

Value

Arguments

Author