HurdleGEE_l: Generalized Estimating Equations, Logistic component of...

In lalondetl/GMM:

Description

This function calculates the Generalized Estimating Equations (GEE) parameter estimates and standard errors for the logistic component of a hurdle model for longitudinal excess zero count responses with independent working correlation structure, based on Dobbie and Welsh (2001). Responses are treated as binary indicators of 0 count values, therefore the probability modeled is that of a zero count. Data must be organized by subject, and an intercept term is assumed for both the logistic and truncated count components of the model. The function outputs a list with parameter estimates betaHat and parameter covariance estimate covEst, along with estimated probabilities pHat and residuals r_l.

Usage

HurdleGEE_l(y, subjectID, N, X_l)

Arguments

y	The vector of response counts, ordered by subject, time within subject.
subjectID	The vector of subject ID values for each response.
N	The number of subjects.
X_l	The design matrix for all covariates in the logistic component of the model, including an intercept.

Examples

HurdleGEE_l()

lalondetl/GMM documentation built on May 30, 2019, 11:40 p.m.