EP1614066A2

EP1614066A2 - Financial product pricing system

Info

Publication number: EP1614066A2
Application number: EP03812638A
Authority: EP
Inventors: Guillaume Blacher
Original assignee: Reech Capital PLC
Current assignee: Fidelity Information Services LLC
Priority date: 2002-12-06
Filing date: 2003-12-05
Publication date: 2006-01-11
Also published as: WO2004053756A2; FR2848315A1; US20080133253A1; AU2003302789A1; US8065220B2; WO2004053756A8

Description

FINANCIAL PRODUCTS PRICE CALCULATION SYSTEM

The invention relates to a system for calculating the price of financial products.

Financial products, and in particular derivative products, are subject to very strict regulations which oblige financial institutions and commercial companies using them to calculate their prices according to a certain number of assumptions which involve variables specific to the characteristics of the products considered and variables specific to the financial markets to which these products belong. These prices must be calculated for valuation purposes and also to be included in the regulatory Risk reports.

Financial, in particular VaR ("Value at Risk"). These price calculations ("PRICING") must be updated very frequently, generally daily, which results in very heavy information processing.

In addition, many new financial products appear regularly. The analysis of each new product in order to be able to calculate the price, the VAR, etc. represents a very heavy work and usually requires the development of a specific price calculation model. This model must then be integrated into the computer processing chains used for the analysis, processing, valuation and calculation of risks. These systems are often different.

The invention aims to provide a system which makes it possible to considerably simplify the operations for calculating the price of a financial product.

To this end, the subject of the invention is a system for calculating the price of a financial product, comprising interface means, data storage means, calculation means and data processing means, remarkable system. in that: a) the interface means comprise means for entering into said system identification and description data of said product, said data comprising: ai) contextual data of said product comprising at least one valuation currency and at at least one underlying, a2) data characteristic of said product comprising a set of events and flows associated with said product, b) the data processing means comprise means for generating from said identification and description data a developed timetable in which an event and / or flow relating to said product is associated with each date, c) the data processing means furthermore comprise means for interpreting said timetable to generate: d) a table of variables of said product from said events and / or flows, c2) for each date of said developed schedule, a function for calculating said product price as a function of at least one of said product variables, d) the interface means comprise means for entering a list of market variables associated with said product and generated by a market analysis, and that, e) the calculation means comprise means for calculating, for each of the scenarios / market conditions and for each of said dates, the values of the product variables given the market variables and means for calculating said product price as a function of said calculated values of the product variables.

According to one embodiment, the data processing means comprise means for generating a compact script containing all the data necessary for calculating the price of the product.

According to another embodiment, the means for entering the identification and description data of said product comprise means for entering said data in the form of a compact script.

According to another embodiment, the means for entering the identification and description data of said product include input windows into which said contextual data and said characteristic data can be entered separately.

According to another characteristic of the invention, the data processing means also comprise means for verifying the interpretation of said schedule.

According to yet another characteristic of the invention, said calculation means comprise: e1) means for calculating, for each of the scenarios / market conditions and for each of said dates, the value of each of said market variables, e2) means for calculating, for each of the scenarios / market conditions and for each of said dates, the value of the product variables as a function of said values of market variables, e3) means for calculating said price as a function of said values of variables of product in all scenarios / market states.

Other characteristics and advantages of the invention will emerge from the description which follows of an embodiment of the invention illustrated by the appended drawings in which:

- Figure 1 is a block diagram illustrating the basic means used by the system according to the invention to describe a financial product and calculate the price of this product;

- Figure 2 is a block diagram illustrating the means and data necessary for the calculation of a financial product by the system according to the invention; - Figure 3 is a block diagram illustrating the process of calculating a financial product using the system according to the invention;

- Figures 4 and 5 are tables illustrating the storage means necessary for the calculation process of Figure 3;

- Figures 6 to 8 illustrate windows for entering information necessary for the description according to the invention of a financial product. Referring to FIG. 1, the calculation of the price of a financial product using the system according to the invention uses the following means and data:

- a calculation tool proper ("PRICER") 1; - product variables 2 which are a function of the nature of the financial product considered (for example SWAP, CALL, etc.);

- market hypotheses 3 known to specialists in financial techniques (for example normal, lognormal, “mean reverting”, multi-factor, etc.), which depend, among other things, on the underlying asset 4 (for example stock, bond , exchange rate, credit) attached to the product under consideration; these market hypotheses make it possible to determine the different scenarios or market conditions and the way in which these scenarios will be generated and modeled, each scenario corresponding to a set of possible values of market variables, which is used to calculate the price of the financial product; - calculation means comprising numerical resolution methods known to specialists in financial techniques (for example trees

6, integration 7, equation to partial derivatives PDE 8, Monte-Carlo 9,) which make it possible to calculate a product price from product variables and market hypotheses.

According to FIG. 2, the system allowing a user to arrive at the calculation of the price of a financial product involves first of all a description of the product. This requires first of all the input by a user, thanks to the system's interface means: - contextual data of the product, namely at least one valuation currency (allowing to define all the conventions of the market or the place concerned) and one or more underlyings (for example, yield curve, "Equity", exchange rate, etc.), namely the market quantities used to define the product; this contextual data indicates which market variables will be involved in calculating the price of the product and for which market hypotheses are chosen;

- data characteristic of the financial product under consideration. This data includes a schedule of events and flows associated with the product.

For this purpose a user can: - either enter a compact script 10 of the type represented in FIG. 8, which contains all the information (contextual data and characteristic data of the product) and which will then be separated into different objects each containing these two types of data,

- either enter directly in separate windows the contextual data on the one hand and the characteristic data of the product on the other hand, as illustrated by the entry screen in Figure 6. From this description and identification data of the product, the data processing means of the system are used to generate at 13 a developed timetable, that is to say an exhaustive series of dates D1, D2 Dn to each of which is associated an event and / or a flow relating to the product (table T1).

A table of objects T2 and a table of product variables T3 are constructed at 14 and 15 respectively by means of an interpreter ("PARSER").

Table T2 is an intermediate table which allows the definition and construction, for each date considered D1, D2 Dn, of one or more trees of evaluation representing a writing in tree of the function Y _t = f (t, X | ... x _n) Y _t - *,) in which:

- Y _t represents the current value of the product variable (s), - 1 represents time, - x * ... x _n represent the value of market variables, market variables which will be identified in 17.

Table T3 is a list of product variables and the values associated with each of these variables, values which are updated by the evolution tree (s) of table T2. At 16, the system uses verification means to ensure the consistency of the data in table T2.

In 17, a market analysis is carried out allowing the construction in 18 of a table of “WORLD” T4 objects. This table consists of a list of the additional market information that is required on each of the dates D1 to Dn to calculate the price or evaluate the product.

These “WORLD” objects are defined on the basis of a few basic elements, four of which are particularly important:

- the “spot” (stocks, commodities, energy, indices, etc.),

- the exchange rate, - the interest rate preferably represented by the discount factors. It can also be zero coupon rates, Libor, swap, etc.

- counterparty default information, for example: has a bond defaulted? It can also be a rate spread ("SPREAD") representing the credit risk.

The market variables being identified at this stage and the table T4 constructed, the construction of the evaluation trees of the table T2 is therefore successful.

From the data obtained as described with reference to FIG. 2, the calculation tool 1 ("PRICER") calculates the price by applying one of the digital resolution methods 5.

These numerical financial resolution methods (for example trees 6, integration 7, equation with partial derivatives PDE 8, Monte-Carlo 9), moreover conventional and well known to finance specialists, have the purpose of:

- simulate or explore the possible values of market variables, - calculate the expected or future value of product variables. The block diagram in FIG. 3 illustrates the numerical resolution of the problem of calculating the price of the product. At 20, depending on the digital resolution method 5 considered, the contextual data used in them, which were obtained as described with reference to FIG. 2, and the number of product variables are acquired. .

In 21, the digital resolution means generate, according to the market assumptions considered, the values of the market variables at each date D1,

D2 Dn of the schedule as well as each scenario established according to these assumptions. As shown in Figure 4, each date and scenario / market state corresponds to a table Tvvm of values of market variables.

In 22, the digital resolution means calculate, for each date and scenario / state considered on the market, the values of the product variables. As Figure 5 shows, for each date and scenario / state considered in the market, there is a Tvp table of the values of the product variables. In 23, the digital resolution means finally formulate a product price as a function of all the calculated values of the product variables.

An exemplary embodiment of the mechanisms making it possible to identify and describe a financial product by its contextual data and its characteristic data as indicated in 10 and 12, to construct the table T1 of FIG. 2, will now be described in more detail with regard to the Figures 6 to 8 which represent input windows for the product to be described.

These mechanisms make it possible to describe any structured financial product, regardless of its underlying and the structure or characteristics of its flows. These mechanisms have: - a syntax defining the type of structure of sentences which are accepted and understood,

- a dictionary of predefined words which are "understood",

- the ability to accept new words, if they are properly defined and introduced. In the following example, a convertible bond will be defined. First, the relevant market variables, namely the currency and its "rate" curve, and the relevant "transferable value", here the Dax, are introduced in windows 30 and 31 of FIG. 6.

It is necessary to describe the amount that the bond will pay at maturity if it is not converted. For this purpose, the term "Redempt" is introduced in 32, this term designating the amount transferred at maturity. A numerical value is given to it in 33, here 100.

Similarly, the “Coupon” and the “ConvPrice” are introduced in 34 and 35, and their respective numerical value in 36 and 37. In order to properly define the product, it is necessary to introduce the concept of “conversion value” ( "Conversion ratio"). This is done at 38 by indicating that "Conv_Ratio (x)" is equal to "100 * (x) / ConvPrice" (window 39).

It is then advisable to describe the product, ie the flows which it will generate and the conditions of this generation if necessary. In the hypothesis considered, the convertible bond pays from any starting point (window 40), for five years (window 41), with an annual frequency (window 42), a coupon (window 43).

After five years (window 44), the bond pays its redemption value (window 45). The bond conversion characteristic is expressed by the fact that at any time ("Od", window 46), during the five years (window 47), the value of the product is the maximum of the value of the product and its conversion value (window 48):

Convert = max (convert, Conv_Ratio (dax)). These mechanisms thus make it possible to describe in a very simple way a convertible bond with particularly unusual characteristics.

The system according to the invention generates and can then display, using the interface means, as shown in FIG. 7, the product flows, from the data entered in the form of a pre-established format. This allows a user to verify that the expected flows are well represented and captured by the system.

Finally, the data processing means allow the generation by the system of a script (FIG. 8), that is to say of a code describing exactly the characteristics of the product and containing all the information necessary for the calculation of the product price. The script, shown in Figure 8, can be exchanged between all stakeholders to describe and calculate the price of the product.

As a variant, all the information, that is to say the contextual data and the characteristic data of the product under consideration, can be entered by a user in the form of a compact script of the type represented in FIG. 8, which the calculator dissociates into different objects, each containing contextual data and characteristic data of the product. The description of the product by windowing, leading to the generation of a script capable of being exchanged between and interpreted by different computers, as described above, or directly in the form of a script, requires use by the user predefined words that are part of a dictionary that can be enriched if necessary; and a syntax defining the type of sentence structure that are accepted and understood.

The syntax rules and the predefined vocabulary are given in the appendix and an example of the use of this vocabulary and these syntax rules is presented below. Thus the definition of the value of the product given in window 48:

Convert = max (convert, Conv_Ratio (dax)) is a variable assignment using a function whose name is predefined "max", which is applied to two parameters:

- the first parameter "convert" is a variable, - the second parameter "Conv_Ratio (dax)" uses a definition (of macro function type) given in windows 38 and 39 which means that the expression "Conv_Ratio (dax) "Must be replaced by the expression" 100 * dax / ConvPrice ", in which" dax "is a variable and" ConvPrice "is an expression which must in turn be replaced by the numeric value" 6500 "due to the definition which is made of it in windows 35 and 37. The definition of window 48 uses a reserved vocabulary word "max", defined as being a function name. The set of reserved function names is defined in the grammar as follows: FUNCTION_NAME = "CONT"

| "MAX" | "MIN"

| "MAXINDEX" | "MININDEX" | "SORT"

| "LOG" | "EXP" | "ABS" | "INT" | "TERM" "SPOT"

"FX"

"DF"

"LIBOR"

"SWAPRATE"

"LEVEL"

"ACCRUEDCOUPON"

"ACCRUALFACTOR"

"Black-Scholes"

("FIX_DATE" | "FIXDATE")

("START_PERIOD" | "STARTPERIOD")

("END_PERIOD" | "ENDPERIOD")

("PAY_DATE" | "PAYDATE")

"GTC"

"START"

"END"

"NOW";

The definition of window 48 also uses the following syntax rules: - definition of an assignment: assignment = ID "=" expression;

- definition of a function: function = (FUNCTION_NAME | auxTab_or_array) "(" exprList ")";

- definition of a list of expressions: exprList = expression {"," expression};

- definition of an expression: expression = sum; sum = product {("+" | "-") product}; product = power {("*" | "/") power}; power = sign {" ^Λ "sign}; sign = {"+" | "-"} atom; atom = function | ID | NUMCONST | "(" expression ")"; We see here that the definition of an expression is given by rules depending successively on each other and using for the last one the definition of a function. It is therefore possible in any expression to call a function which itself uses expressions, also using functions and so on recursively.

Claims

1. System for calculating the price of a financial product, comprising interface means, data storage means, calculation means and data processing means characterized in that: a) the interface means comprise means for entering into said system identification and description data of said product, said data comprising: ai) contextual data of said product comprising at least one valuation currency and at least one underlying, a2) characteristic data of said product comprising a set of events and flows associated with said product, b) the data processing means comprise means for generating from said identification and description data a developed schedule (T1) in which at each date is associated an event and / or flow relating to said product, c) the data processing means further comprises means for interpreting said schedule er to generate: d) a table of variables (T3) of said product from said events and / or flows, c2) for each date of said developed schedule, a function for calculating said product price as a function of at least one of said variables of product, d) the interface means comprise means for entering a list of market variables (T4) associated with said product and generated by a market analysis, and that, e) the calculation means comprise means for calculating, for each of the scenarios / market conditions and for each of said dates, the values of the product variables being given the market variables and means for calculating said product price as a function of said calculated values of product variables.

2. System according to claim 1 characterized in that the data processing means comprise means for generating a compact script containing all the data necessary for calculating the price of the product.

3. System according to claim 2 characterized in that the means for entering the identification and description data of said product comprise means for entering said data in the form of a compact script.

4. System according to claim 1 characterized in that the means for entering the identification and description data of said product include input windows into which said contextual data and said characteristic data can be entered separately

5. System according to any one of the preceding claims, characterized in that the data processing means also comprise means for checking the interpretation of said schedule.

6. System according to any one of the preceding claims, characterized in that said calculation means comprise e1) means for calculating, for each of the scenarios / market conditions and for each of said dates, the value of each of said market variables, e2) means for calculating, for each of the scenarios / market conditions and for each of said dates, the value of the product variables as a function of said values of market variables, e3) means for calculating said price as a function of said values of product variables in all scenarios / market conditions.

7. A system for calculating the price of a financial product according to claim 6 characterized in that the data storage means comprise means for storing in the form of tables said values of the market variables (Twm).

8. System for calculating the price of a financial product according to any one of the preceding claims, characterized in that the data storage means comprise means for storing said schedule (T1), said calculation functions (T2) in the form of tables , said product variables (T3), market variables (T4) and said values of product variables (Tvp).