invHessvar: Calculates asymptotic variance using Hessian.

In david-borchers/hmltm: Line transect estimation with hidden Markov availability models

Calculates asymptotic variance using Hessian.

Description

Calculates asymptotic variance of a statistic derived from a hsltm fit, using the inverse of the Hessian matrix and estimated derivative of the statististic with respect to the parameter vector. The derivative is approximated using function splinefun.

Usage

invHessvar(stat, hmmlt, nx4spline = 50, doplot = FALSE)

Arguments

stat	name of statistic to calculate. Valid statistics are "p0" for estimated probability at perpendicular distance zero, "p" for mean estimated detection probability over all perpendicular distances, "invp" for the inverse of mean estimated detection probability over all perpendicular distances, "esw" for estimated effective strip width, and "invesw" for estimated inverse of effective strip width.
hmmlt	object output by fit.hmltm.
nx4spline	is number of points at which to evaluate function for spline interpolation.
doplot	if TRUE, plots the spline approximation to the statistic as a function of each parameter.

See Also

hmltm.stat for calculation of derived statistics, and splinefun for approximation of function derivatives.

david-borchers/hmltm documentation built on Oct. 29, 2023, 9:07 p.m.