Generate Poisson TDMM simulation data — data.gen.tdmm.pois • TDMM

Generate Poisson counts for TDMM examples. The default uses repeated measurements and a subject-specific random intercept; random_intercept = FALSE generates independent observations with two predictors.

Usage

data.gen.tdmm.pois(n.subject = 200, n.time = 20, sigma2.b = 0.5,
  seed = NULL, random_intercept = TRUE)

Arguments

n.subject: Number of subjects. If omitted, the default is 200 with a random intercept and 2200 without one. An explicitly supplied value is used in either case.
n.time: Number of observations per subject when random_intercept = TRUE; unused otherwise.
sigma2.b: Random-intercept variance when random_intercept = TRUE; unused otherwise.
seed: Optional random seed.
random_intercept: A single logical value. TRUE generates repeated observations on a common time grid with a subject-specific random intercept. FALSE generates one independent observation per subject with time drawn uniformly from 0 to 1.

Details

With random_intercept = FALSE, x1 is standard Normal and x2 = u + 0.5*time + e, where u is standard Normal and e is Normal with standard deviation 0.5, independently drawn for each subject. The log mean is beta0(time) + x1*beta1(time) + x2*beta2(time), where beta0(t) = 0.25 + 0.35*sin(2*pi*t), beta1(t) = 0.35*cos(2*pi*t), and beta2(t) = 0.25 + 0.20*t.

Value

A data frame with subject.ID, time, y, and x1; the independent-observation design also has x2.

Examples

d <- data.gen.tdmm.pois(n.subject = 5, n.time = 4, seed = 1,
  random_intercept = TRUE)
d.independent <- data.gen.tdmm.pois(n.subject = 10,
  random_intercept = FALSE, seed = 1)