DSBlog

Comments (1)

The missions of the U.S. financial and intelligence communities have never been so closely aligned as they are today. Dennis Blair, former Director of National Intelligence (DNI), recently stated that “the primary near-term security concern of the United States is the global economic crisis and its geopolitical implications.” In further evidence of this alignment, the President’s Daily Brief has been augmented since early 2009 by a classified daily Economic Intelligence Brief (aka the Butterfly Brief), and the CIA has been actively recruiting ex-Wall Streeters.

Over the last decade or so, Christina Ray has had a foot in each of these two very different worlds. As an experienced Wall Street trader and risk manager, she came from the realm of the probable, in which historical data and sophisticated stochastic methods are used to value securities and estimate future risk. As a member of Digital Sandbox’s Board of Advisors, however, she later grew to appreciate the realm of the plausible as well.

Now Ray, a senior managing director for Market Intelligence at Omnis Inc., has published Extreme Risk Management (McGraw Hill; June 2010), which provides an overview of current best practice in the financial community but also describes an alternative path to actionable intelligence using a connectivist approach. Judging by the growing interest in the book from the risk management community, it’s an approach whose time has come. According to Dr. Robert Mark, author and Managing Partner of Black Diamond Risk Enterprises, “The ideas that flow throughout Extreme Risk Management are nicely crafted to provide the missing ingredients for a paradigm shift toward the next generation of practical and theoretically sound risk management methodologies.”

Recent events such as the global financial crisis have shown that some of the most extreme scenarios are more plausible than formerly believed. Some observers – most notably Nassim Taleb and Malcolm Gladwell – have written eloquently about ‘black swan’ events and ‘tipping points,’ but provide few tangible methods for identifying, quantifying or warning of such events. Extreme Risk Management addresses analytical methods employed in both the financial and intelligence spaces that might be used toward this end.

As Ray observes, “Over the last three decades or so, sophisticated financial modeling has been almost exclusively statistical in nature. The ready availability of massive amounts of historical market data has fueled the creation of valuation and risk measurement models built on concepts such as association, correlation, and likelihood.”

But she goes on to point out, as the old saying goes, that “Correlation is not causation. The alternative to a statistical model is a causal model that explicitly creates an alternative worldview, one in which cause and effect are modeled in temporal or logical order.”

As the author further explains:

“All these [financial] models create explicit forecasts, that is, estimates of expected and possible future scenarios for a security or portfolio of securities. Most often, these forecasts are based on the assumption that the future market behavior is well represented by the past.

However, this stochastic approach implies a worldview that ignores causality in favor of correlation. In this world, it doesn’t matter whether gold prices increased because interest rates decreased or vice versa. It also doesn’t matter whether the price of a utility stock and an airline stock are directly related in some fashion or whether, instead, they are both driven by a common dependence on fuel prices. This world is a supremely efficient world as well: all prices reflect information immediately, and that information is transmitted instantaneously around the globe.

However, intuition belies these notions. Traders and portfolio managers know that events drive prices. Catalysts such as the release of an economic indicator or an earnings report drive prices, and chain reactions precipitated by an important event can take a finite amount of time to propagate.

This alternative world is one in which plausibility rather than probability is modeled. The consequences and likelihood of events that have never before occurred but that can be reasonably anticipated (as a consequence of other events) are included in the quantitative models. Such modeling is the forte of the intelligence community and those responsible for national security, who must create metrics and construct solutions for threats that have never before occurred.”

Ray makes the case that it’s now easier to create alternative models that measure plausibility. As computer scientist and philosopher Judea Pearl famously said: “Causality has been mathematized”.

Coming in Part 2: Implications for national policy and security, and the benefits of ‘MARKINT.’

––––––––––––––––

Digital Sandbox is the leader in public safety risk management, providing analytic tools and information products to government agencies and large enterprises, for optimizing risk-based strategic, policy, and budgetary decisions.