nLogLike: Negative log-likelihood function

In bmcclintock/momentuHMM: Maximum Likelihood Analysis of Animal Movement Behavior Using Multivariate Hidden Markov Models

Negative log-likelihood function

Description

Negative log-likelihood function

Usage

nLogLike(
  optPar,
  nbStates,
  formula,
  bounds,
  parSize,
  data,
  dist,
  covs,
  estAngleMean,
  circularAngleMean,
  consensus,
  zeroInflation,
  oneInflation,
  stationary = FALSE,
  fullDM,
  DMind,
  Bndind,
  knownStates,
  fixPar,
  wparIndex,
  nc,
  meanind,
  covsDelta,
  workBounds,
  prior = NULL,
  betaCons = NULL,
  betaRef,
  deltaCons = NULL,
  optInd = NULL,
  recovs = NULL,
  g0covs = NULL,
  mixtures = 1,
  covsPi,
  recharge = NULL,
  aInd
)

Arguments

optPar	Vector of working parameters.
nbStates	Number of states of the HMM.
formula	Regression formula for the transition probability covariates.
bounds	Named list of 2-column matrices specifying bounds on the natural (i.e, real) scale of the probability distribution parameters for each data stream.
parSize	Named list indicating the number of natural parameters of the data stream probability distributions
data	An object momentuHMMData.
dist	Named list indicating the probability distributions of the data streams.
covs	data frame containing the beta model covariates (if any)
estAngleMean	Named list indicating whether or not to estimate the angle mean for data streams with angular distributions ('vm' and 'wrpcauchy').
circularAngleMean	Named list indicating whether to use circular-linear or circular-circular regression on the mean of circular distributions ('vm' and 'wrpcauchy') for turning angles. See fitHMM.
consensus	Named list indicating whether to use the circular-circular regression consensus model
zeroInflation	Named list of logicals indicating whether the probability distributions of the data streams are zero-inflated.
oneInflation	Named list of logicals indicating whether the probability distributions of the data streams are one-inflated.
stationary	FALSE if there are time-varying covariates in formula or any covariates in formulaDelta. If TRUE, the initial distribution is considered equal to the stationary distribution. Default: FALSE.
fullDM	Named list containing the full (i.e. not shorthand) design matrix for each data stream.
DMind	Named list indicating whether fullDM includes individual- and/or temporal-covariates for each data stream specifies (-1,1) bounds for the concentration parameters instead of the default [0,1) bounds.
Bndind	Named list indicating whether DM is NULL with default parameter bounds for each data stream.
knownStates	Vector of values of the state process which are known prior to fitting the model (if any).
fixPar	Vector of working parameters which are assumed known prior to fitting the model (NA indicates parameters is to be estimated).
wparIndex	Vector of indices for the elements of fixPar that are not NA.
nc	indicator for zeros in fullDM
meanind	index for circular-circular regression mean angles with at least one non-zero entry in fullDM
covsDelta	data frame containing the delta model covariates (if any)
workBounds	named list of 2-column matrices specifying bounds on the working scale of the probability distribution, transition probability, and initial distribution parameters
prior	A function that returns the log-density of the working scale parameter prior distribution(s)
betaCons	Matrix of the same dimension as beta0 composed of integers identifying any equality constraints among the t.p.m. parameters.
betaRef	Indices of reference elements for t.p.m. multinomial logit link.
deltaCons	Matrix of the same dimension as delta0 composed of integers identifying any equality constraints among the initial distribution working scale parameters.
optInd	indices of constrained parameters
recovs	data frame containing the recharge model theta covariates (if any)
g0covs	data frame containing the recharge model g0 covariates (if any)
mixtures	Number of mixtures for the state transition probabilities
covsPi	data frame containing the pi model covariates
recharge	recharge model specification (only used for hierarchical models)
aInd	vector of indices of first observation for each animal

Value

The negative log-likelihood of the parameters given the data.

Examples

## Not run: 
# data is a momentuHMMData object (as returned by prepData), automatically loaded with the package
data <- example$m$data
m<-example$m
Par <- getPar(m)
nbStates <- length(m$stateNames)

inputs <- momentuHMM:::checkInputs(nbStates,m$conditions$dist,Par$Par,m$conditions$estAngleMean,
          m$conditions$circularAngleMean,m$conditions$zeroInflation,m$conditions$oneInflation,
          m$conditions$DM,m$conditions$userBounds,
          m$stateNames)

wpar <- momentuHMM:::n2w(Par$Par,m$conditions$bounds,list(beta=Par$beta),
        log(Par$delta[-1]/Par$delta[1]),nbStates,m$conditions$estAngleMean,
        m$conditions$DM,m$conditions$Bndind,
        m$conditions$dist)

l <- momentuHMM:::nLogLike(wpar,nbStates,m$conditions$formula,m$conditions$bounds,
     inputs$p$parSize,data,inputs$dist,model.matrix(m$conditions$formula,data),
     m$conditions$estAngleMean,m$conditions$circularAngleMean,inputs$consensus,
     m$conditions$zeroInflation,m$conditions$oneInflation,m$conditions$stationary,
     m$conditions$fullDM,m$conditions$DMind,
     m$conditions$Bndind,m$knownStates,unlist(m$conditions$fixPar),
     m$conditions$wparIndex,covsDelta=m$covsDelta,workBounds=m$conditions$workBounds,
     betaRef=m$conditions$betaRef,covsPi=m$covsPi)

## End(Not run)

bmcclintock/momentuHMM documentation built on Oct. 26, 2022, 1 a.m.