On data-based optimal stopping under stationarity and ergodicity

DOI10.3150/12-BEJ439zbMATH Open1273.62192arXiv1307.5976OpenAlexW3103370543MaRDI QIDQ358137FDOQ358137

Authors: Harro Walk, Michael Kohler 

Publication date: 16 August 2013

Published in: Bernoulli (Search for Journal in Brave)

Abstract: The problem of optimal stopping with finite horizon in discrete time is considered in view of maximizing the expected gain. The algorithm proposed in this paper is completely nonparametric in the sense that it uses observed data from the past of the process up to time $-n+1$, $ninmathbbN$, not relying on any specific model assumption. Kernel regression estimation of conditional expectations and prediction theory of individual sequences are used as tools. It is shown that the algorithm is universally consistent: the achieved expected gain converges to the optimal value for $noinfty$ whenever the underlying process is stationary and ergodic. An application to exercising American options is given, and the algorithm is illustrated by simulated data.

Full work available at URL: https://arxiv.org/abs/1307.5976

Recommendations

• On the optimal stopping with incomplete data
• scientific article; zbMATH DE number 4052699
• On the optimal stopping values induced by general dependence structures
• scientific article; zbMATH DE number 6166689
• Publication:4734631
• Optimal stopping under model uncertainty: randomized stopping times approach
• Statistical inference for a finite optimal stopping problem with unknown transition probabilities
• Optimal stopping problems in Lévy models with random observations
• Optimal Stopping for Processes with Independent Increments, and Applications

zbMATH Keywords

ergodicitypredictionstationaritynonparametric regressionAmerican optionsuniversal consistency

Mathematics Subject Classification ID

Nonparametric regression and quantile regression (62G08) Derivative securities (option pricing, hedging, etc.) (91G20) Applications of statistics to actuarial sciences and financial mathematics (62P05) Optimal stopping in statistics (62L15)

Cites Work

• Prediction, Learning, and Games
• A distribution-free theory of nonparametric regression
• Ergodic theorems. With a supplement by Antoine Brunel
• Title not available (Why is that?)
• Title not available (Why is that?)
• Valuing American options by simulation: a simple least-squares approach
• THE GARCH OPTION PRICING MODEL
• Title not available (Why is that?)
• Monte Carlo algorithms for optimal stopping and statistical learning
• A review on regression-based Monte Carlo methods for pricing American options
• Optimal stopping of Markov processes: Hilbert space theory, approximation algorithms, and an application to pricing high-dimensional financial derivatives
• Nonparametric inference for ergodic, stationary time series
• Pricing Bermudan options by nonparametric regression: optimal rates of convergence for lower estimates
• NONPARAMETRIC KERNEL‐BASED SEQUENTIAL INVESTMENT STRATEGIES
• Nonparametric nearest neighbor based empirical portfolio selection strategies

Cited In (4)

• Weakly universally consistent static forecasting of stationary and ergodic time series via local averaging and least squares estimates
• Optimal stopping for image smoothing based on kernel density estimation
• On the optimal stopping with incomplete data
• Nonparametric learning for impulse control problems -- exploration vs. exploitation

This page was built for publication: On data-based optimal stopping under stationarity and ergodicity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q358137)