asci

approximate spatial correlation implementation

Usage

asci(a, b, a_x, a_y, b_index, scale_fact, err = 1e-8)

Arguments

a

vector of values for observation set \(A\)

b

vector of values for observation set \(B\)

a_x

x-axis coordinates for observation set \(A\)

a_y

y-axis coordinates for observation set \(A\)

b_index

spatial index for observation set \(B\) (see build_index)

scale_fact

scale factor for negative exponential kernel

err

error term in approximation, s.t. for all observation in sets \(A\) and \(B\) only consider observation pairs \(a\), \(b\) such that:

\(e^{\frac{-d(a, b)}{s}} \geq \epsilon\) (or rephrased: \(d(a, b) \leq -ln(\epsilon) * s\))

where:

• \(d\) is euclidian distance between observation \(a\) and \(b\)

• \(s\) is the scale factor

• \(\epsilon\) is the error term

Value

(approximate) spatial correlation metric \(c\)

Examples

set.seed(8)

a_n <- 128
b_n <- 256

a <- rpois(a_n, runif(1, 0, 32))
a_x <- rnorm(a_n)
a_y <- rnorm(a_n)

b <- rpois(b_n, runif(1, 0, 32))
b_x <- rnorm(b_n)
b_y <- rnorm(b_n)

b_idx <- build_index(b_x, b_y)

asci(a, b, a_x, a_y, b_idx, 1e-1)
#> [1] 0.03093763