Simulate interval Beta-regression data (flexible)
simulation_DATA.beta(
n,
p,
s = min(5L, p),
beta_size = 1,
a0 = 0,
X_dist = c("gaussian", "t", "bernoulli"),
corr = c("indep", "ar1", "block"),
rho = 0,
block_size = 5L,
df = 5,
prob = 0.5,
active_idx = NULL,
phi = 20,
mechanism = c("jitter", "quantile", "mixed"),
mix_prob = 0.5,
delta = 0.05,
delta_low = NULL,
delta_high = NULL,
alpha = 0.1,
alpha_low = NULL,
alpha_high = NULL,
na_rate = 0,
na_side = c("random", "left", "right"),
centerX = FALSE,
scaleX = FALSE,
seed = NULL
)list with X, Y, Y_low, Y_high, mu, beta, a0, phi, info, settings.
Sample size and number of predictors.
Number of active (nonzero) coefficients.
Scalar (alternating ±) or numeric vector of length greater then equal s.
Intercept (logit scale).
Distribution for X: "gaussian", "t", or "bernoulli".
Correlation structure: "indep", "ar1", or "block".
AR(1) correlation or within-block correlation.
Block size when corr = "block" (default 5).
Degrees of freedom for X_dist = "t" (default 5).
Success prob for X_dist = "bernoulli" (default 0.5).
Optional integer vector of active feature indices (length s). If NULL, uses 1:s.
Precision parameter: scalar, length-n vector, or function (mu, X) -> length-n.
Interval mechanism per row: "jitter", "quantile", or "mixed".
Probability of jitter when mechanism = "mixed".
Symmetric jitter half-width (scalar / vector / function).
Asymmetric jitter widths (override delta if set).
Miscoverage for quantile intervals (scalar / vector / function).
Asymmetric miscoverage (override alpha if set).
Fraction of rows with a missing bound (default 0).
Which bound to drop: "left", "right", or "random".
Whether to center/scale X before returning.
RNG seed.