dsdive.generator.matrix.uniformized: Compute infinitesimal generator matrix for a homogeneous,...

In jmhewitt/dsdive: Discrete Space Models for Dive Trajectories of Animals

Description

Compute infinitesimal generator matrix for a uniformized Continuous time Markov chain (CTMC). See Rao and Teh (2013) for an applied definition and use of infinitesimal generator matrices.

Usage

dsdive.generator.matrix.uniformized(
  depth.bins,
  beta,
  lambda,
  s0,
  rate.uniformized
)

Arguments

depth.bins	n x 2 Matrix that defines the depth bins. The first column defines the depth at the center of each depth bin, and the second column defines the half-width of each bin.
beta	Directional preference model parameters. See dsdive.tx.params for more details.
lambda	Diving rate model parameters. See dsdive.tx.params for more details.
s0	dive stage for which matrix should be computed
rate.uniformized	uniformization rate, for standardizing transition rates between states

References

Rao, Vinayak, and Yee Whye Teh. "Fast MCMC sampling for Markov jump processes and extensions." The Journal of Machine Learning Research 14.1 (2013): 3295-3320.

Examples

data('dive.sim')
attach(dive.sim)
attach(dive.sim$params)

r.unif = max(outer(lambda, 2 * depth.bins$halfwidth, '/'))

m = dsdive.generator.matrix.uniformized(
  depth.bins = depth.bins, beta = beta, lambda = lambda, s0 = 2, 
  rate.uniformized = r.unif)

detach(dive.sim$params)
detach(dive.sim)

jmhewitt/dsdive documentation built on May 29, 2020, 5:18 p.m.