sampleData: Simulate data with binary or time-to-event outcome

In riskRegression: Risk Regression Models and Prediction Scores for Survival Analysis with Competing Risks

Simulate data with binary or time-to-event outcome

Description

Simulate data with binary outcome and 10 covariates.

Usage

sampleData(n,outcome="competing.risks",
formula= ~ f(X1,2)+f(X2,-0.033)+f(X3,0.4)+f(X6,.1)+f(X7,-.1)+f(X8,.5)+f(X9,-1),
          intercept=0)
sampleDataTD(n,n.intervals=5,outcome="competing.risks",
formula= ~ f(X1,2)+f(X2,-0.033)+f(X3,0.4)+f(X6,.1)+f(X7,-.1)+f(X8,.5)+f(X9,-1))

Arguments

n	Sample size
outcome	Character vector. Response variables are generated according to keywords: "binary" = binary response, "survival" = survival response, "competing.risks" = competing risks response
formula	Specify regression coefficients
intercept	For binary outcome the intercept of the logistic regression.
n.intervals	sampleDataTD only: the maximum number of episodes in which the covariates are updated.

Details

For the actual lava::regression parameters see the function definition.

Value

Simulated data as data.table with n rows and the following columns: Y (binary outcome), time (non-binary outcome), event (non-binary outcome), X1-X5 (binary predictors), X6-X10 (continous predictors)

Author(s)

Thomas A. Gerds <[e-mail]>

See Also

lvm

Examples

set.seed(10)
sampleData(10,outcome="binary")
sampleData(10,outcome="survival")
sampleData(10,outcome="competing.risks")

riskRegression documentation built on Sept. 8, 2023, 6:12 p.m.