Weighted-least-squares hunting — hunt

Returns the hunted signal h as a function

Usage

hunt_wls(
  wls_alt_method,
  resids,
  X,
  X.cols = 1:ncol(X),
  trim.outlier = TRUE,
  arg.wls_alt_method = NULL,
  predict_fun_alt = stats::predict
)

Arguments

wls_alt_method: Function with signature wls_alt_method(y, X, w, ...) that returns a fitted alternative model \(\hat{g} \in \mathcal{G}\) by minimizing \(\sum_i w_i (y_i - g(x_i))^2\). The returned object must support predict_fun_alt(g, X) for evaluation.
resids: Residuals (i.e., negative scores) of length n from the null model.
X: Covariates of dim n x p.
X.cols: Subset of covariates to hunt. (Default: 1:ncol(X))
trim.outlier: If TRUE, outliers in \(\hat{h}(X)\) will be trimmed from the hunted \(\hat{h}\) using Tukey's IQR rule.
arg.wls_alt_method: Named list of additional arguments passed to wls_alt_method (default to NULL).
predict_fun_alt: Function with signature predict_fun_alt(fit, X) returning a numeric vector of predictions from the alternative-model fit. Default stats::predict.

Value

A function h with signature h(X).