evaluator: Function to evaluate spatial quantiles

In cepp: Context Driven Exploratory Projection Pursuit

Description

This provides an objective function whose minimization yields the spatial quantiles.

Usage

evaluator(n, p)

Arguments

n	The number of rows in the data
p	The number of columns in the data

Details

Returns another function suitable for passing to an optimizer like nlm or trust.

Value

A function that should be passed to an optimizer.

Author(s)

Mohit Dayal

References

P. Chaudhuri. "On a geometric notion of quantiles for multivariate data." Journal of the American Statistical Association, 91(434):862-872, 1996.

Examples

x <- rnorm(500)
dim(x) <- c(250,2)
ev <- evaluator(250,2)
##The Spatial Median
trust(ev, parinit=c(median(x[1,]), median(x[2,])), u=c(0,0),
      rinit=0.5, rmax=2e5, samp = x)
##Quantile for vector (0.2,0.3)
trust(ev, parinit=c(median(x[1,]), median(x[2,])), u=c(0.2,0.3),
      rinit=0.5, rmax=2e5, samp = x)

Example output

Loading required package: trust
$value
[1] 301.5399

$gradient
[1] -8.647249e-12 -1.531967e-12

$hessian
           [,1]       [,2]
[1,] 176.571684  -5.100188
[2,]  -5.100188 148.483856

$argument
[1] 0.05795514 0.07271672

$converged
[1] TRUE

$iterations
[1] 6

$value
[1] 272.1432

$gradient
[1] -4.121148e-13 -1.421085e-13

$hessian
          [,1]      [,2]
[1,] 170.99491  -1.82928
[2,]  -1.82928 137.38850

$argument
[1] 0.3421160 0.5412897

$converged
[1] TRUE

$iterations
[1] 6

cepp documentation built on May 2, 2019, 3:44 p.m.