Confidence Intervals for Spatial Half-Life — spurhalflife • spatialunitroot

Constructs confidence sets for the half-life of spatial processes based on Mueller and Watson (2024). The half-life is the distance at which spatial correlation drops to 1/2.

Usage

spurhalflife(
  formula,
  data,
  coords = NULL,
  level = 0.95,
  q = 15,
  nrep = 1e+05,
  latlong = FALSE,
  normdist = FALSE,
  seed = NULL
)

Arguments

formula: A formula specifying the variable, e.g., y ~ 1.
data: A data frame containing the variables.
coords: Coordinate specification: a formula (e.g., ~ s_1 + s_2), a matrix, or NULL to auto-detect s_* variables.
level: Confidence level (between 0 and 1). Default: 0.95.
q: Number of low-frequency weighted averages. Default: 15.
nrep: Number of Monte Carlo replications. Default: 100000.
latlong: Logical; are coordinates latitude/longitude?
normdist: Logical; if TRUE, return CI in normalized distance units (fractions of maximum distance). If FALSE (default), return CI in metres (if latlong) or original coordinate units.
seed: Optional seed for reproducibility.

Value

An object of class "spur_halflife" with components:

ci_lower: Lower bound of the confidence interval.
ci_upper: Upper bound of the confidence interval (Inf if unbounded).
max_dist: Maximum pairwise distance in original units.
level: Confidence level used.
normdist: Whether CI is in normalized units.
q: Number of low-frequency components.
nrep: Number of Monte Carlo replications.
call: The matched call.

References

Mueller, U. K. and Watson, M. W. (2024). "Spatial Unit Roots and Spurious Regression." Econometrica, 92, 1661–1695.

Examples

if (FALSE) { # \dontrun{
data(chetty)
spurhalflife(am ~ 1, data = chetty, coords = ~ s_1 + s_2, latlong = TRUE)
} # }