ContrivedData: Simulated Point-Referenced Spatial Data

In lbaole17/krige.ext: Geospatial Kriging with Metropolis Sampling (Experimental Version)

Description

These data are a simulated point-referenced geospatial data that serve to provide a clean example of a kriging model. There are 500 observations with coordinates located on a unit square.

Usage

data(NY_subset)
     

Format

The ContrivedData dataset has 500 observations and 5 variables.

y

The outcome variable. Its true population functional form is y_s=0+1 x_{1s}+2 x_{2s}+ω_{s}+ε_{s}. The true variance of ω is σ^2=0.5 and of ε is τ^2=0.5. The decay term that shapes spatial correlation levels is φ=2.5.

x.1

A predictor with a standard uniform distribution.

x.2

A predictor with a standard normal distribution.

s.1

Coordinate in eastings for each observation, distributed standard uniform.

s.2

Coordinate in northings for each observation, distributed standard uniform.

Examples

#Examine Data
summary(ContrivedData)

#Initial OLS Model
contrived.ols<-lm(y~x.1+x.2,data=ContrivedData);summary(contrived.ols)

lbaole17/krige.ext documentation built on Sept. 29, 2020, 1:52 a.m.