In a previous article we discussed what problems models and their assumptions introduced which could have precipitated the subprime crisis. Now let’s look at the analysts or the modelers themselves.
We recently encountered a fairly experienced risk manager who was very well-versed in quantitative methods, unlike a vast majority of "check-box" risk managers. His question was, what new information does entropy give him that his current sophisticated Bayesian models and analyses don’t? Business analytics professionals may have the same question.
The venerable World Economic Forum recently published their 5th annual "Global Risks" report. While clearly this is a very timely analysis, we see a few problems with their report.
Traditional risk analysis involves developing what are known as "ordinal scoring" scales. For example, this requires getting executives to answer questions like "what is the likelihood that the next major database upgrade will be delayed", "what is the impact of the hurricane season on the availability of medicines for our troops" etc.