Least Square Regression Methods for Bermudan Derivatives and Systems of Functions
Least square regression methods are Monte Carlo methods to solve non-linear problems related to Markov processes and are widely used in practice. In these methods, first we choose a system of functions to approximate value functions. So one of questions on these methods is what kinds of systems of functions one has to take to get a good approximation. In the present paper, we will discuss on this problem.
KeywordsComputational finance Option pricing Malliavin calculus Least square regression methods
- 3.Castaing C, Valadier M (1977) Convex analysis and measurable multifunctions. Lecture notes in mathematics, vol 580. Springer, Berlin/New YorkGoogle Scholar
- 6.Kusuoka S, Morimoto Y (2014) Stochastic mesh methods for Hörmander type diffusion processes. In: Kusuoka S, Maruyama T (eds) Advances in mathematical economics, vol 18. Springer, Tokyo Heidelberg New York Dordrecht London, pp 61–99Google Scholar