Fat Tails Quantified and Resolved: A New Distribution to Reveal and\n Characterize the Risk and Opportunity Inherent in Leptokurtic Data

2011/10/29 by Lawrence R. Thorne, Thorne, Lawrence R., L. R. Thorne
Computer Science · Mathematics · #Advanced Clustering Algorithms Research #Advanced Statistical Methods and Models #Statistical Methods and Bayesian Inference

paper · pdf · doi:10.48550/arxiv.1110.6553

Abstract

I report a new statistical distribution formulated to confront the infamous,\nlong-standing, computational/modeling challenge presented by highly skewed\nand/or leptokurtic ("fat- or heavy-tailed") data. The distribution is\nstraightforward, flexible and effective. Even when working with far fewer data\npoints than are routinely required, it models non-Gaussian data samples, from\npeak center through far tails, within the context of a single probability\ndensity function (PDF) that is valid over an extremely broad range of\ndispersions and probability densities. The distribution is a precision tool to\ncharacterize the great risk and the great opportunity inherent in fat-tailed\ndata.\n

Related

• Statistical Consequences of Fat Tails: Real World Preasymptotics, Epistemology, and Applications
• A useful family of fat-tailed distributions
• A Peer-based Model of Fat-tailed Outcomes
• Where Do Thin Tails Come From?
• Fat Tailed Distributions in Catastrophe Prediction