cf.smcu: Straddle Maximum Contour Uncertainty criterion

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

Straddle Maximum Contour Uncertainty criterion

Description

straddle MCU with a specified variance weight

Usage

cf.smcu(objMean, objSd, gamma = 1.96)

Arguments

objMean	predicted mean response
objSd	posterior standard deviation of the response
gamma	weight on the variance

Details

compute the UCB criterion with constant weight: gamma*s(x) - |f(x)|

Author(s)

Mike Ludkovski

References

Mike Ludkovski, Kriging Metamodels and Experimental Design for Bermudan Option Pricing Journal of Computational Finance, 22(1), 37-77, 2018

X.Lyu, M Binois, M. Ludkovski (2020+) Evaluating Gaussian Process Metamodels and Sequential Designs for Noisy Level Set Estimation <https://arxiv.org/abs/1807.06712>

See Also

[osp.seq.design]

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