crwN2ll: -2 * log-likelihood for CTCRW models

Description

This function is designed for primary use within the crwMLE model fitting function. But, it can be accessed for advanced R and crawl users. Uses the state-space parameterization and Kalman filter method presented in Johnson et al. (2008).

Usage

crwN2ll(theta, fixPar, y, x, loctype, delta, a1.y, a1.x, P1.x, P1.y, lonAdj,
  mov.mf, err.mfX, err.mfY, stop.mf, n.errX, n.errY, n.mov, stopMod, driftMod,
  prior, need.hess, constr = list(lower = -Inf, upper = Inf))

Arguments

theta

parameter values.

fixPar

values of parameters held fixed (contains NA for theta values).

latitude locations.

longitude loations.

loctype

vector with 1 for unobserved locations, and 0 for observed locations.

delta

time difference to next location.

a1.y

initial state value for latitude.

a1.x

initial state value for longitude.

P1.x

intial state covariance matrix for latitude.

P1.y

inital state covariance matrix for longitude.

lonAdj

= 1/cos(y*pi/180) as described in Johnson et al. (2008) for polar coords. = 1 for non-polar coords.

mov.mf

Movement covariate data.

err.mfX

longitude error covariate data.

err.mfY

latitude error covariate data.

stop.mf

stopping covariate.

n.errX

number or longitude error parameters.

n.errY

number of latitude error parameters.

n.mov

number or movement parameters.

stopMod

Logical. indicates wheteher a stop model is specified.

driftMod

Logical. inicates whether a drift model is specified.

prior

Function of theta that returns the log-density of the prior

need.hess

Whether or not the Hessian will need to be calculated from this call

constr

Named list giving the parameter constraints

Value

-2 * log-likelihood value for specified CTCRW model.

Details

This function calls compiled Fortran code which can be viewed in the src directory of the crawl library.

References