PALMA: an HPCN-enabled stochastic system for dynamic ALM
by
Gaetano Zanghirati
Dipartimento di Matematica, Univeristà di Ferrara
Coauthors: Flavio Cocco (Prometeia s.r.l., Bologna), Giuseppe Paruolo (CINECA, Bologna)

The availability of an Asset & Liability Management (ALM) system based on a dynamic and stochastic economic and financial scenario simulation exploiting HPCN technology can significantly increase the user's competitiveness in the today's market. Such a tool implement a mathematical and statistical model aimed at forecasting the future expected revenues of the financial firm and at measuring the volatility of the expected revenues. Using HPCN resources makes it possible to increase the precision of risk estimation, optimise capital allocation and enhance pricing and portability of financial products.

The dynamics of firm's revenues is the result of the interaction of two factors: the dynamics of financial and credit markets, which are the main source of uncertainty for the firm, and the strategies of the firm with respect to the set of variables under management's control (asset and liabilities composition, pricing).

The main goal of an ALM system is to support the management in improving the performance of the firm by optimising capital allocation among different business opportunities and by properly pricing products and services of the firm with respect to the level of risk undertaken in different lines of business.

We report the promising results of the parallelisation strategy adopted to implement such a stochastic dynamic ALM code obtained on a 128-proc Cray T3E, under the European PST project named PALMA (Parallel Asset and Liability MAnagement). Anyway the code is easily portable on other, possibly etherogeneous, high performance computing platforms.

Date received: February 10, 1999

Copyright © 1999 by the author(s). The author(s) of this document and the organizers of the conference have granted their consent to include this abstract in Atlas Conferences Inc. Document # cacq-04.