dot-fast_plausibility_contour: Fast plausibility_contour calculation when pred_fun =...

In mhuiying/scp: Spatial Conformal Prediction

Fast plausibility_contour calculation when pred_fun = Krige_fun and (dfun = "residual2" or "std_residual2")

Description

Internal function provides the conformal prediction interval for spatial location s0 given partial obserations s1,...,sm, Y1,.., Ym when the square of the standard kriging residuals are used as the non-conformality measures, and weights are provided.

Usage

.fast_plausibility_contour(Y_aug, w_aug, Q, std = TRUE)

Arguments

Q	an n+1 \times n+1 inverse of estimated covariance matrix
s0	prediction location
s	an n \times d matrix or data-frame with d coordinates of the n data locations.
Y	a vector with n data values.
w	weights for the non-conformity measures.
alpha	significance level. Defaults to 0.05.

Value

A vector of lower and upper bounds of the conformal prediction interval.

mhuiying/scp documentation built on May 4, 2022, 11:35 p.m.