ExpDistMat: Exponential distance matrix

In RozetaSimonovska/SDPDmod: Spatial Dynamic Panel Data Modeling

Exponential distance matrix

Description

This function calculates the (negative) exponential distance matrix, with a given cutoff distance and a positive exponent value.

Usage

ExpDistMat(distMat, distCutOff = NULL, expn = 0.01, mevn = FALSE)

Arguments

distMat	distance matrix
distCutOff	cutoff distance. Default = the maximal value from the distance matrix.
expn	positive exponent, default = 0.01
mevn	logical, default FALSE. If TRUE, max-eigenvalue normalization is performed.

Details

W is an nxn matrix with elements w_{ij}, i, j = 1,..n, where w_{ij}=e^{-\alpha d_{ij}}, if 0 <= d_{ij} < D and w_{ij}=0, if d_{ij} > D or i = j. D is the distance cutoff point (maximum radius of influence), d_{ij} is the distance between spatial units i and j, and \alpha is the positive exponent (e.g. \alpha= 0.01, 0.02,...).

Value

W	spatial weights matrix (Default, not normalized)

Author(s)

Rozeta Simonovska

Examples

data(gN3dist) ##distance in meters
W1    <- ExpDistMat(distMat = gN3dist, distCutOff = 100000)
dist2 <- gN3dist/1000 ##in km
W2    <- ExpDistMat(distMat = dist2, distCutOff = 100, expn = 0.02)
W2nor <- ExpDistMat(distMat = dist2, 100000, 0.001, mevn = TRUE)

RozetaSimonovska/SDPDmod documentation built on April 14, 2024, 9:40 p.m.