loglik.pt.1re: Loglikelihood of Poisson-Tweedie generalized linear mixed model with random intercept

Description

Evaluates the loglikelihood of a Poisson-Tweedie generalized linear mixed model with random intercept, using the adaptive Gauss-Hermite quadrature rule.

Usage

loglik.pt.1re(beta, D, a, Sigma, y, X, Z, id, offset = NULL, GHk = 5,
  tol = 9.88131291682493e-324, GHs = NULL)

Value

The loglikelihood value obtained using a Gauss-Hermite quadrature approximation with GHk quadrature points.

Arguments

beta: Vector of regression coefficients
D: Dispersion parameter (must be > 1)
a: Power parameter (must be < 1)
Sigma: A matrix with the variance of the random intercept
y: Response vector (discrete)
X: Design matrix for the fixed effects
Z: Design matrix for the random effects
id: Id indicator (it should be numeric)
offset: Offset term to be added to the linear predictor
GHk: Number of quadrature points (default is 5)
tol: Tolerance value for the evaluation of the probability mass function of the Poisson-Tweedie distribution
GHs: Quadrature points at which to evaluate the loglikelihood. If NULL (default value), the GH quadrature points are computed internally

Author

Mirko Signorelli

References