Plot local quantile calibration for PLQR fits (experimental)

For each target, the weighted share of observations below the quantile threshold, sum_i w_ij 1{Y_i <= D_i' theta + g_i}, is plotted against the nominal level tau with a reference band tau +/- z * sqrt(tau (1 - tau) / n_eff). Shares outside the band flag local miscalibration of the supplied theta_eval or first stages — the applied reading of the quantile moment condition.

Usage

plot_calibration(object, residuals = NULL, level = 0.95, ...)

Arguments

object: A plqr_result with the weight matrix W stored (the default).
residuals: Optional numeric vector of length n of quantile residuals; used when the fit does not store the indicator.
level: Confidence level of the reference band (default 0.95).
...: Ignored; reserved for future options.

Value

A ggplot object.

Details

The below-threshold indicator is stored on the fit; alternatively, supply residuals (Y - D' theta - g, length n), whose non-positive entries define the indicator.

Lifecycle

Experimental. PLQR functionality is outside the CATE and regression workflow focus of the initial public release. Its API and inference interfaces may change in later releases.

Examples

if (requireNamespace("ggplot2", quietly = TRUE)) {
  demo <- plqr_demo_data(n = 90, p = 2, p_d = 2, tau = 0.5, seed = 1)
  df <- demo$data
  fit <- plqr_dnn(
    X = as.matrix(df[paste0("X", 1:2)]),
    D = as.matrix(df[paste0("D", 1:2)]),
    g = as.numeric(df$g_oracle),
    mD = as.matrix(df[paste0("mD", 1:2)]),
    Y = as.numeric(df$Y),
    theta_eval = demo$truth$theta_true, tau = 0.5,
    X0 = rbind(c(0, 0), c(0.5, 0.5)), k = 1L, s = 8L, se = FALSE
  )
  plot_calibration(fit)
}