fit.mnr: Fit Multivariate Outcome Model

In zrmacc/MNR: Multivariate Normal Regression

Description

Fits a regression model in which a multivariate normal random vector is observed for each subject. Regression models are specified using a list of numeric matrices, one for each column of Y.

Usage

fit.mnr(Y, X, sig = 0.05, maxit = 10, eps = 1e-06, report = T)

Arguments

Y	Outcome matrix.
X	List of model matrices, one for each outcome.
sig	Significance level, for confidence intervals.
maxit	Maximum number of parameter updates.
eps	Minimum acceptable improvement in log likelihood.
report	Report fitting progress? Default is FALSE.

Value

An object of class mnr containing the estimated regression parameters, covariance matrix, the information matrix for regression parameters, and the residuals.

Examples

## Not run: 
# See rMNR or vignette for data generation
M = fit.mnr(Y=Y,X=X,eps=1e-8);
# Coefficients
coef(M);
# Outcome covariance
vcov(M,type="Outcome");
# Information matrix
vcov(M,type="Information");
# Residuals
resid(M);

## End(Not run)

zrmacc/MNR documentation built on May 17, 2019, 8:47 p.m.