Professional LHWR Model Report — lhwr

Print a structured report for an LHWR fitted model. The report is designed for applied analysis: it combines the model call, calibration settings, a global Gaussian or binomial baseline, LHWR diagnostics, local coefficient summaries, optional local inference, optional diagnostic F-tests, and optional residual or collinearity diagnostics.

Usage

lhwr_report(
  object,
  include_global = TRUE,
  include_inference = TRUE,
  include_f_tests = TRUE,
  include_moran = !is.null(moran),
  include_collinearity = !is.null(collinearity),
  local_inference = NULL,
  moran = NULL,
  collinearity = NULL,
  alpha = 0.05,
  adjust = "BH",
  digits = 4,
  ...
)

Arguments

object: An LHWR model object returned by lhwr().
include_global: Logical. If TRUE, include the global linear or logistic baseline stored in object$lm when available.
include_inference: Logical. If TRUE, include exploratory local coefficient inference when available or computable from the fit.
include_f_tests: Logical. If TRUE, include diagnostic F-tests when stored in the fit or computable from the fit.
include_moran: Logical. If TRUE, include the object supplied through moran. The function does not compute Moran's I automatically.
include_collinearity: Logical. If TRUE, include the object supplied through collinearity. The function does not compute local collinearity diagnostics automatically.
local_inference: Optional object returned by local_inference(). If NULL and include_inference = TRUE, the function attempts to compute it from object.
moran: Optional object returned by moran.lhwr() or moran.hgwr().
collinearity: Optional object returned by local_collinearity().
alpha: Significance level used when local inference must be computed.
adjust: Multiplicity adjustment used when local inference must be computed. Passed to local_inference().
digits: Number of significant digits used in printed numeric values.
...: Unused.

Value

Invisibly returns an object of class "lhwr_report" containing the tables printed in the report.

Details

The function does not refit the model. Local inference is computed only from quantities already stored in the fitted object, and therefore requires a fit created with diagnostics = "full". Moran and local collinearity diagnostics can be passed through the moran and collinearity arguments when they have already been computed separately.

Examples

set.seed(7)
H <- simulate_H2(80, r_max = 3)
Z <- H$Z
x1 <- rnorm(80)
x2 <- rnorm(80)
y <- 1 + 2 * x1 - 0.5 * x2 + rnorm(80, sd = 0.3)
dat <- data.frame(y, x1, x2)
fit <- lhwr(y ~ x1 + x2, data = dat, Z = Z, bw = 35,
            adaptive = TRUE, diagnostics = "full", f_tests = TRUE)
lhwr_report(fit, include_moran = FALSE, include_collinearity = FALSE)