Centralised stochastic simulation process

Abstract

A method of simulating a local system which is in interaction with other local systems within a global environment. The method includes generating, at a central location, a collection of global sets of values, each representing a possible future global state of the global environment at a future time, and the collection representing a variety of possible future global states. The method also includes transmitting the collection of global sets of values to the local systems. The local systems receive some of the collection, select a local set of parameters, and perform calculations on that data. The collection of values is used as a predictive simulation of the local system.

Description

[0001]This is a continuation application of Ser. No. 10 / 168,013 filed Jan. 28, 2003 (now U.S. Pat. No. 7,542,881, issued Jun. 2, 2009), which is a U.S. national stage application of PCT / FR00 / 02055 filed Jul. 17, 2000, which claims priority benefit to French application number 00 / 06113 filed May 11, 2000.TECHNICAL FIELD[0002]The invention relates to the technical field of “Industrial Risk Management” and more specifically to:[0003]Statistic / probabilistic prediction processes of the probability law on statistic distribution of variations that indicate future behaviour and optimisation of productivity parameters for a number of remote local industrial systems subjected to a common global environment, both random and evolutionary in nature.[0004]Simulation processes, i.e. which use a predetermined sampling of probable future scenarios of environment state variables.[0005]The invention relates specifically to the following cases:[0006]Where a vast number of decentralised industrial sites...

AI Technical Summary

Benefits of technology

[0123]Only perform simple or less problematic calculations concerning information technology and specialist personnel at the vast number of local industrial systems.

Problems solved by technology

In contracts, non-quantifiable risks may be legal, fraud or media related risks.

However, the method is not sufficient enough to outline the risk structure of complex random systems.

This is insufficient to describe the reality of risk behaviour for a complex random system.

This is also insufficient to describe the reality of risk behaviour for a complex random system.

e) In different cases where we would want to predict random changes such as variations in petrol or share prices rather than predict a risk, too many factors come into play that are themselves too unpredictable and difficult to model with mathematical tools.

However, these methods are unsuitable for general risk analysis of industrial systems in view of common, especially regulation, criteria.

Within the global financial world, market risk is one that the investor faces due to his lack of knowledge concerning future evolution of common state variables that are market basics, such as share prices, interest rates, currency rates and so on.

However, these different risk approaches known as “sensitive”:Are specific to a financial product type.Rely on specific distribution hypotheses.Are not easy to interpret.Often underestimate the global loss risk on a portfolio which is actually of interest to the investor.Do not assess the probability level of risk occurrence.Do not consider temporal forecast.

The basic problem with risk analysis statistic / probabilistic processes (and especially VaR calculation processes) is therefore to predict distribution of the probability law and distribution of variations for a numerical productivity indicator (loss in the case of VaR).

Despite its superior position over the “delta normal” methods, the “delta gamma” method is generally unsuitable, especially to assess portfolio risk that includes options reaching maturity or exotic derivative products such as “knock-out” options.

The methods of the previous work as outlined above do not allow a satisfactory probabilistic prediction of future behaviour and / or the optimisation of productivity parameters for a number of remote local industrial systems subjected to a common global environment to be performed.

In particular, they do not allow good VaR conditions of a number of financial sites managing share portfolios to be simultaneously calculated.

However, transmitted values are not probable multi-value samplings of environment variables but simple parameters (see above).

The drawback to this method is that it requires very strict base hypotheses (conditional normality, linearity, stationariness) that are not acceptable, in the most part, for financial instruments.

Thus, VaR calculation according to the variance-covariance matrix method produces errors in the case of portfolios with “gamma” or “convexity” (namely inversions of convexity or discontinuities) instruments, such as options, structured accounts, “mortgage-backed securities” and so on.

In general, this is insufficient to describe environments with complex risk factors.B) The classical simulation method, and in particular, the Monte-Carlo simulation method, is considered by the majority of specialists to be the “purest”, the most global and forward-looking in its conception.

Nevertheless, it requires complex choices of assessment models and stochastic processes that generate correctly chosen random samplings.

This is very complex given the difficulty in assessing certain instruments and their constant development.

The availability of large information systems is therefore necessary and expensive.

In such a way that it is not widely acceptable to the majority of financial institutions or industrial sites that would wish for and be interested in its implementation in order to reduce their risk.C) The major inconvenient of the historical analysis by simulation method is its extreme sensitivity to data quality.

Unfortunately, the amount of calculations involved in its use is prohibitive.

Further, the historical VaR are inaccurate and must be completed using appropriate “stress tests”.

Consequently, and according to the historical method of the previous work, each site must be equipped with a costly information system and a large team to implement it.

In such a way that it may also not be accessible to the majority of financial institutions or industrial sites that would wish for or have an interest in its implementation in order to reduce their risk.

However, this does not allow for a global probabilistic view (probability law in the statistic distribution of variations for a numerical behaviour indicator) and is limited to a small number of parameters for each site.

Images