treeDivide.BW: Create a Bouchard-Warin equi-probable grid of regression...

In mludkov/mlOSP: Machine Learning and Regression Monte Carlo Algorithms for Optimal Stopping

Create a Bouchard-Warin equi-probable grid of regression sub-domains

Description

Create a Bouchard-Warin equi-probable grid of regression sub-domains

Usage

treeDivide.BW(grid, curDim, model, test.paths)

Arguments

grid	dataset of x-values
curDim	dimension of the grid
model	a list containing all model parameters. In particular must have model$nBins defined
test.paths	testing paths to predict along as well

Details

Recursively sort along each of the d-coordinates At the end do local linear regression at each leaf This is a recursive algorithm! first column is reserved for the y-coordinate (timingValue) It's safest if nrows(grid) is divisible by model$nBinsmodel$dim

mludkov/mlOSP documentation built on April 29, 2023, 7:56 p.m.