Use of Peyton Jones’ Contract Descriptive Language to Evaluate Different Value Processes

Abstract

Abstract Contracts play a major role in financial markets. Financial contracts vary in different aspects from one to another and as such, they need proper formalization and mechanism for valuation. Peyton Jones have introduced a contract descriptive language that can be used for both representation and valuation of financial contracts. This language has become popular among the research community and hence many improvements and applications have been made with this language since its introduction. This research aims to explore different processing activities (value processes) that can be applied to Peyton Jones’ contract descriptive language. By doing so the language becomes more powerful from its functionalities. In this research, we mainly explore two such processing activities. Those are, generating a calendar for contractual obligations and valuating stochastic processes using Monte Carlo simulation. Creating a model to generate a calendar for a contract written in Peyton Jones’ contract descriptive language helps users to track the timeline of the contract. We introduce a model that has the capability of generating a calendar for a given contract. This model consists of calendar definition, a set of combinators for calendars and a set of evaluation semantics for the conversion from a contract to the calendar. Peyton Jones have proposed a valuation model for their contract descriptive language. They used a lattice valuation model to calculate the present value of future cash flows of the contracts. They have claimed that instead of this lattice valuation model, other valuation techniques can be used. Our aim is to support this claim by creating a model to valuate contracts in Peyton Jones’ contract descriptive language using Monte Carlo simulation. A simple Monte Carlo simulation method described by Boyle for contracts is used for valuation of contracts. The Monte Carlo simulation and evaluation semantics of contracts are implemented with stochastic processes.