XploRe Help : genglm

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library:	glm
See also:	glmest genmultlo

Quantlet:		genglm
Description:		genglm generates data from a GLM model.

Link:

Tutorial: GLM in XploRe

Usage:	`{y,mu} = genglm(code, x, b {,opt})`
Input:
	`code`	text string, the short code for the model (e.g. "bilo" for logit or "noid" for ordinary PLM).
	`x`	n x k matrix, design.
	`b`	k x 1 vector, coefficients.
	`opt`	optional, a list with optional input. "glmopt" can be used to set up this parameter. The order of the list elements is not important.
	`opt.phi`	nuisance parameter, usually the dispersion parameter.
	`opt.wx`	scalar or n x 1 vector, the binomial index vector, only considered for binomial models. If not given, set to 1.
	`opt.off`	scalar or n x 1 vector, offset. If not given, set to 0.
	`opt.pow`	optional, power for power link.
	`opt.nbk`	scalar, extra parameter k for negative binomial distribution. 1/k is supposed to be a positive integer. If not given, set to 1 (geometric distribution).
Output:
	`y`	n x 1 vector, response.
	`mu`	n x 1 vector, regression.

Note:

The Gamma models "gacl" and "gapow" are not yet implemented.

Example:

library("glm") n = 100 b = 2|(-1) x = normal(n,rows(b)) opt = glmopt("wx",round(uniform(n).*5)+1) {y,mu} = genglm("bilo",x,b,opt)

Result:

y[i]'s are pseudo-random variables with distribution
Binomial(wx[i],mu[i]) where mu[i]=1/(1+exp(-x[,i]*b).

Author: M. Mueller, 20010228