biLocal: Estimate the local correlation between two variables when the...

In hotneim/lgde: Implements the Locally Gaussian Density Estimator

Description

Estimate the local correlation between two variables when the marginals are standard normal.

Usage

biLocal(data, grid, h)

Arguments

data	The data matrix, one row per observation. The marginals are assumed to be standard normal. They can be transformed using the transLocal-function.
grid	The grid where the local correlation will be estimated.
h	The vector of bandwidths. Length must be 2.

Value

A list with 4 elements: $data is the data, $grid is the grid, $par.est is a vector of the estimated local correlations, and $f.est are the local likelihood density estimates there.

Examples

data <- cbind(rt(100, df = 10), rt(100, df = 10))
grid <- matrix(c(-1, -1, 0, 0, 1, 1), ncol = 2, byrow = TRUE)
trans <- transLocal(data = data, grid = grid)

estimates <- biLocal(data = trans$transformed.data,
                     grid = trans$transformed.grid,
                     h = c(1, 1))

hotneim/lgde documentation built on May 17, 2019, 4:52 p.m.