validMCBinom_Types: Generalized Method of Moments Valid Moment Combinations for...

In lalondetl/GMM:

Description

This function calculates the values of valid moment combinations for two-step Generalized Method of Moments with user-defined types of time-dependent covariates, applied to longitudinal data with count (0-n) outcomes. It is assumed that the count represents the number of events from n identical trials, and that n is equal for all subjects and times. This is modeled similarly to a Logistic Regression for Binomial responses. It allows for unbalanced longitudinal data, meaning subjects can be observed for different numbers of times. The function returns a vector "types" indicating validity of different moments conditions.

Usage

validMCBinom_Types(ymat, subjectIndex, Zmat, Xmat, covTypeVec, betaI, T, Tmax,
  Count)

Arguments

ymat	The matrix of responses, ordered by subject, time within subject. The first column is the number of successes, the second the number of failures.
Zmat	The design matrix for time-independent covariates.
Xmat	The design matrix for time-dependent covariates.
covTypeVec	The vector indicating the type of each time-dependent covariate.
betaI	The current or initial estimates of the model parameters.
T	The number of time points for subject i.
Tmax	The maximum number of times of observation among all subjects.
Count	A vector of running counts of the number of valid moment conditions for all subjects.

Examples

validMCBinom_Types()

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