Introduction

Risk management depends heavily on analytics to generate information for decision making. It is no surprise, therefore, that computing has played a major role in risk management from the very beginning. In the past, risk measurement and reporting systems have largely been segregated by risk type – with separate systems for measuring credit, market and liquidity risk – though in some cases integrated market and credit engines have been developed. This approach is appealing because the systems are usually easier to implement when separate. It was also workable in an earlier age when it was tolerable for risk to be managed in silos. Recent developments have changed this dynamic significantly – not only do risk management systems need to integrate measurements across risk types due to needs like stress testing, but they also need to more readily integrate with transaction and finance systems to enable better Risk-Based Pricing (RBP) and Risk-Adjusted Performance Management (RAPM).

One unintended but favorable side-effect of segregated systems was that each computer system was not overly burdened by data processing requirements. The need for more integration has therefore posed a new and significant challenge to data processing technologies. Fortunately, we are in a period of tremendous innovation in data processing – data storage and processing technology have progressed to a point where these advanced needs are becoming eminently feasible. This chapter will first review the information needs of risk management, followed by a broad overview of data processing technology.