ordinal.data {simml} | R Documentation |
A function for ordinal categorical response data generation.
Description
ordinal.data
generates ordered category response data (with p covariates and a treatment variable).
Usage
ordinal.data(n = 400, p = 10, R = 11, delta = 1, s = "nonlinear",
sigma = 0)
Arguments
n |
sample size. |
p |
dimension of covariates. |
R |
number of response levels in y |
delta |
magnitude of "main" effect (i.e., "nuisance" effect) of the covariates; a large delta means a larger "nuisance" variance. |
s |
type of the treatment-by-covariates interation effect ("linear" or "nonlinear") |
sigma |
noise sd in the latent variable representation |
Value
y |
a n-by-1 vector of treatment outcomes. |
A |
a n-by-1 vector of treatment indicators. |
X |
a n-by-p matrix of pretreatment covariates. |
SNR |
the "signal" (interaction effect) to "nuisance" (main effect) variance ratio (SNR) in the canonical parameter function. |
true.beta |
the true single-index coefficient vector. |
delta |
magnitude of "main" effect. |
s |
type of the treatment-by-covariates interation effect. |