n_eff: Effective sample size

In geostan: Bayesian Spatial Analysis

Effective sample size

Description

An approximate calculation for the effective sample size for spatially autocorrelated data. Only valid for approximately normally distributed data.

Usage

n_eff(n, rho)

Arguments

n	Number of observations.
rho	Spatial autocorrelation parameter from a simultaneous autoregressive model.

Details

Implements Equation 3 from Griffith (2005).

Value

Returns effective sample size n*, a numeric value.

Source

Griffith, Daniel A. (2005). Effective geographic sample size in the presence of spatial autocorrelation. Annals of the Association of American Geographers. Vol. 95(4): 740-760.

See Also

sim_sar, aple

Examples

n_eff(100, 0)
n_eff(100, 0.5)
n_eff(100, 0.9)
n_eff(100, 1)

rho <- seq(0, 1, by = 0.01)
plot(rho, n_eff(100, rho),
     type = 'l',
     ylab = "Effective Sample Size")

geostan documentation built on April 3, 2025, 10:04 p.m.