Simulate outcome

Simulates outcome vector.

Usage

.simulate_outcome(family, x = NULL, beta = NULL, n = NULL, factor = 1)

Arguments

family

character "gaussian", "binomial", "poisson", or "cox"

x

numeric \(n \times p\) matrix

beta

numeric \(p\)-dimensional vector

n

integer or NULL

factor

non-negative scalar (default: factor=1) for multiplying the linear predictor (to increase or decrease the signal strength)

Value

Returns an \(n\)-dimensional outcome vector.

See also

Use simulate() to simulate a predictor matrix, an effect vector, and an outcome vector.

Examples

# simulate independent outcome
corila:::.simulate_outcome(family = "gaussian", n = 10)
#>  [1]  0.14457084 -0.91107471  0.48129531  0.43516490 -0.04162846 -0.20415504
#>  [7]  0.05402925 -0.76406361 -0.69892937 -0.63471855

# simulate dependent outcome
n <- 10
p <- 20
x <- matrix(rnorm(n * p), n, p)
beta <- rnorm(p)
corila:::.simulate_outcome(family = "gaussian", x = x, beta = beta)
#>  [1] -0.6618132 -0.1856400 -1.3569514 -1.2114130 -2.2969133  0.3021709
#>  [7]  2.3079562  1.8628679 -1.4648566  1.8779476