PCDHess.wf: Calculating the Hessian matrix using Louis's method (1982)

In PCDSpline: Semiparametric regression analysis of panel count data using monotone splines

Description

Calculating the Hessian matrix using Louis's Method under the Gamma frailty non-homogeneous Poisson process model. This is a support function for PCDReg.wf.

Usage

PCDHess.wf(DATA, beta, gamma, nu, order, knots)

Arguments

DATA	use specified data structure.
beta	estimates of regression coefficients.
gamma	estimates of spline coefficients.
nu	estimates of gamma frailty variance parameter.
order	the order of basis functions.
knots	the equally spaced knots.

Details

To obtain the Hessian matrix of the observed likelihood evaluated at the last step output of the EM algorithm.

Value

HESS	Hessian matrix.

References

Louis, T. (1982). Finding the observed information matrix when using the EM algorithm. Journal of the Royal Statistical Society, Series B 44, 226-233.

Yao,B., Wang, L., and He, X. (2014+).Semiparametric regression analysis of panel count data allowing for within-subject correlation.

PCDSpline documentation built on May 2, 2019, 5:14 a.m.