Fast Weighted Identity-Link Binomial Regression, Estimate Only (C++ Backend)

Fits the same identity-link (risk-difference) binomial regression as fast_identity_binomial_regression_cpp (see that page for the full model, boundary-constrained IRLS line search, and interpretation), with each observation's contribution to the log-likelihood and IRLS working weights multiplied by a nonnegative row weight weights_r[i]. Setting all weights to 1 recovers fast_identity_binomial_regression_cpp exactly; this is the backend used when the identity-link model must be fit on bootstrap-reweighted or otherwise weighted data.

Usage

fast_identity_binomial_regression_weighted_cpp(
  X,
  y_r,
  weights_r,
  maxit = 100L,
  tol = 1e-06,
  fixed_idx = NULL,
  fixed_values = NULL,
  warm_start_beta = NULL,
  smart_cold_start = TRUE,
  warm_start_weights = NULL,
  warm_start_fisher_info = NULL,
  estimate_only = FALSE
)

Arguments

X: A numeric matrix of predictors, \(n \times p\).
y_r: A binary (0/1) numeric vector of responses, length \(n\).
weights_r: A nonnegative numeric vector of length \(n\) giving each row's weight.
maxit: Maximum number of Fisher-scoring iterations.
tol: Convergence tolerance.
fixed_idx: Optional integer indices of coefficients to hold fixed rather than estimate.
fixed_values: Optional values to fix the parameters named by fixed_idx at.
warm_start_beta: Optional starting values for coefficients. If provided, smart_cold_start is ignored.
warm_start_weights: Optional initial working weights for the first IRLS iteration.
warm_start_fisher_info: Optional initial Fisher Information matrix for the first IRLS iteration.

Value

A list with the same components as fast_identity_binomial_regression_cpp: b, mu_hat, working_weights, iterations, converged, and fisher_information (all reflecting the weighted log-likelihood).

Usage

Arguments

Value

See also