Monte Carlo methods via a dual approach for some discrete time stochastic control problems

  title={Monte Carlo methods via a dual approach for some discrete time stochastic control problems},
  author={L. Gyurk{\'o} and B. Hambly and J. Witte},
  journal={Mathematical Methods of Operations Research},
We consider a class of discrete time stochastic control problems motivated by a range of financial applications. We develop a numerical technique based on the dual formulation of these problems to obtain an estimate of the value function which improves on purely regression based methods. We demonstrate the competitiveness of the method on the example of a gas storage valuation problem. 
Reinforced optimal control
The reinforced regression method is extended to a general class of stochastic control problems, while considerably improving the method's efficiency, as demonstrated by substantial numerical examples as well as theoretical analysis. Expand
Optimal Switching on a Finite Time Horizon
In this study we investigate the problem of finding the optimal investment strategy in an uncertain market. We consider an investor who can choose whether and when to enter or exit the market in anExpand
A First‐Order BSPDE for Swing Option Pricing
It is shown that the value process solves a first-order non-linear backward stochastic partial differential equation and the set of optimal controls is characterized and a dual minimization problem is derived. Expand
A Convex Optimization Approach to Multiple Stopping: Pricing Chooser Caps and Swing Options
In this current work, we generalize the recent Pathwise Optimization approach of Desai et al to Multiple stopping problems. The approach also minimizes the dual bound as in Desai et al to find theExpand
Least Squares Monte Carlo and Approximate Linear Programming: Error Bounds and Energy Real Option Application
This research supports the use of regress-later LSM rather than this ALP technique to approximately solve intractable Markov decision processes, and provides both theoretical and numerical backing for the usefulness of this solution. Expand
Comparison of least squares Monte Carlo methods with applications to energy real options
This finding suggests the use of LSML in lieu of LSMN when estimating dual bounds on the value of early or multiple exercise options, as well as of related capacity investment and inventory/production policies. Expand
Comparison of Least Squares Monte Carlo Methods with Applications to Energy Real Options
Least squares Monte Carlo (LSM) is a state-of-the-art approximate dynamic programming approach used in financial engineering and real options to value and manage options with early or multipleExpand
Relationship between least squares Monte Carlo and approximate linear programming
It is shown that regress-later LSM is itself an ALP relaxation that potentially corrects this ALP shortcoming and supports using this LSM version rather than ALP on the considered models. Expand
Forest of Stochastic Trees: A Method for Valuing Multiple Exercise Options
The Forest of Stochastic Trees (FOST) method for pricing multiple exercise options by simulation is presented and it is proved the sign of the estimator bias and show that these estimators are consistent for the true option value. Expand
Energy Real Options: Valuation and Operations
This book chapter introduces models for the valuation and operations of energy real options, with an emphasis on approximate dynamic programming methods.


Pathwise Stochastic Optimal Control
  • L. Rogers
  • Mathematics, Computer Science
  • SIAM J. Control. Optim.
  • 2007
This paper approaches optimal control problems for discrete-time controlled Markov processes by representing the value of the problem in a dual Lagrangian form, which opens up the possibility of numerical methods based on Monte Carlo simulation, which may be advantageous in high-dimensional problems or in problems with complicated constraints. Expand
Regression Methods for Stochastic Control Problems and Their Convergence Analysis
The main idea behind the algorithms is to simulate a set of trajectories under some reference measure and to use the Bellman principle combined with fast methods for approximating conditional expectations and functional optimization. Expand
Regression Methods for Stochastic Control Problems
In this paper we develop several regression algorithms for solving general stochastic optimal control problems via Monte Carlo. This type of algorithms is particularly useful for problems withExpand
Monte Carlo Methods for the Valuation of Multiple-Exercise Options
We discuss Monte Carlo methods for valuing options with multiple-exercise features in discrete time. By extending the recently developed duality ideas for American option pricing, we show how toExpand
An analysis of a least squares regression method for American option pricing
Under fairly general conditions, this paper proves the almost sure convergence of the complete algorithm due to Longstaff and Schwartz and determines the rate of convergence of approximation two and proves that its normalized error is asymptotically Gaussian. Expand
A Semi-Lagrangian Approach for Natural Gas Storage Valuation and Optimal Operation
A semi-Lagrangian method for solving the HJB equation for a typical gas storage valuation problem is presented and is able to handle a wide class of spot price models that exhibit mean-reverting seasonality dynamics and price jumps. Expand
Monte Carlo Methods in Financial Engineering
This paper presents a meta-modelling procedure that automates the very labor-intensive and therefore time-heavy and therefore expensive and expensive process of manually computing random numbers and random Variables. Expand
Primal and Dual Pricing of Multiple Exercise Options in Continuous Time
A dual minimization problem for the price of the multiple exercise option in terms of martingales and processes of bounded variation is derived and it is explained how the primal and dual pricing formulas can be applied to compute confidence intervals on the option price via Monte Carlo methods. Expand
Monte Carlo Valuation of American Options
This paper introduces a dual way to price American options, based on simulating the paths of the option payoff, and of a judiciously chosen Lagrangian martingale. Taking the pathwise maximum of theExpand
Dual Representations for General Multiple Stopping Problems
A dual representation for generalized multiple stopping problems, hence the pricing problem of general multiple exercise options is studied, which allows for cashflows which are subject to volume constraints modeled by integer valued adapted processes and refraction periods modeled by stopping times. Expand