sampledata: robspdep generated dataset

In vincnardelli/robspdep: robspdep: Robust Spatial Dependence

robspdep generated dataset

Description

Data is generated considering a 10-by-10 regular square lattice grid where data is generated by a spatial autoregressive process with a strong spatial correlation parameter (rho = 0.7) obtained by filtering a random sample generated by 100 independent standard normal distributions. The outcoming x variable is associated to a Moran test of 0.65 which is highly significant (p-value < 2.2e-16) The variable y is then contaminated with just one single outlier (=50) located in the upper left corner. The new map is now characterized by a much lower value of the Moran test of (0.199) which is not significant (p-value = 0.42). Therefore the presence of a single ouliers totally rubs off the strong spatial correlation structure that was evident before the contamination.

Usage

sampledata

Format

A list containing:

x

uncontaminated variable

y

contaminated variable

w

10-by-10 regular square lattice grid in spdep list w object

vincnardelli/robspdep documentation built on April 11, 2024, 10:30 a.m.