Linear Regression Webinar Assets - Click here

Black Swan Events

Webinar | June 15

Black Swan Events: Measuring and Managing Risk for Rare Events and Economic Uncertainty

Register Now

Reserve your seat for this live webinar to get analyst answers to your risk management questions.

June 15, 2022 10:00 – 11:00AM PDT

(1:00PM Eastern / 5:00PM UTC)

Dr. Christian Smart,
Chief Scientist Galorath

During this webinar, you’ll learn about:

  • Predicting extreme events such as unknown unknowns in the aggregate using historical data.
  • Black swan events, however, cannot be predicted in advance, even though in hindsight they may seem to have been foreseeable.
  • Even though black swan events cannot be predicted, learn how they can be managed.
  • IT risk can be quantified. Many of the extreme examples of IT cost growth and schedule delays are not black swans, but can be predicted in advance by credible analytical techniques.

Risk management for Black Swans, or rare events that have significant negative consequences, is challenging.

Predicting the occurrence of such events is difficult and sometimes impossible. For many events with extreme consequences, there is no typical occurrence. Such phenomena will likely have a few relatively mild outcomes, followed by an extreme one that breaks all previous records.

COVID-19 is a prime example of this phenomenon. The world had not seen such a widespread and deadly pandemic since the great influenza outbreak 100 years ago. The use of traditional statistical methods is not useful in managing risk in these situations. Fat-tailed phenomena do not typically have a well-defined population variance and in many cases not even a population mean! The commonly used ‘normal’ distribution is not applicable in such cases. In this presentation, the use of historical data for managing the risk of extreme events is discussed.

Despite its challenges, much can be leveraged from historical experience in such cases. The use of extreme value theory to the measurement and management of extreme risks is discussed. A key component of many extreme events, such as pandemics, is exponential growth. The importance of swift action to cut the fat tail of extreme events is emphasized. A connection with the development phase of defense projects is discussed.

With many examples of extreme cost growth, the management of risk for national security projects is a tough problem. The use of logistic regression to identify candidates for extreme cost growth is one possible way to help identify potential problem programs before they occur. We discuss the application of extreme value theory, logistic regression, and the importance in cutting the fat tail through mitigation in the risk management of defense projects.