Introduced in 2016, the metalog distributions are most flexible and easiest to use of all probability distributions and have been increasingly widely adopted for practical use. If you have data (expert elicited, empirical, or simulated) and are not sure which continuous probability distribution to use, you are likely best off starting with a metalog. That’s because metalogs are more shape-flexible; faster and easier to fit to data; allow a choice of boundedness (unbounded, semi-bounded, or bounded); have simple closed-form equations; and are easier to simulate. Using freely available spreadsheet or other software, you can typically determine a metalog that’s an excellent fit to your data in less than a minute.
This talk is a one-hour introduction to the metalog distributions that covers the basic concepts and tools that you’ll need to get going. In addition, we’ll show how metalogs can be used for expert elicitation with real-time feedback; show how multiple expert-elicited metalogs can be combined to yield a consensus metalog in closed form; and touch on closed-form Bayesian updating of metalogs when new data becomes available. Come join the fun and add new capabilities to your risk-analysis toolbox.
Tom has combined a decision-professional career with innovations to advance the field. Tom is a founder and Managing Partner of Keelin Reeds Partners, a firm that provides strategy, decision analysis, and education services. Previously, over two decades with the Strategic Decisions Group (SDG), Tom co-developed with his colleagues the definition of “decision quality”, supervised or led dozens of decision analysis or strategy engagements for Fortune 1000 companies, founded SDG’s life sciences practice, and served as head of human resources, board member, and Worldwide Managing Director. In the 1980’s, Tom co-developed and published an award-winning methodology for electric power capacity planning and served as an expert witness on decision quality. In the 1990’s, with his Harvard Business Review article “How SmithKline Beecham Makes Better Resource Allocation Decisions”, Tom introduced portfolio-management methods that are widely used today and that led to significant job creation for decision professionals. In the 2000’s, Tom developed and published the metalog distributions to help increase decision professionals’ productivity. Tom is a Fellow of the Society of Decision Professionals and a founder and director of the not-for-profit Decision Education Foundation. He holds three Stanford degrees: BA in Economics and MS and PhD in Engineering-Economic Systems.