First Significant Digit Distributions in the Credit Crisis

• Activity Number: 174
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Paul Hofmarcher*+ and Florian Löcker
• Companies: WU Wien and Institute for Statistics and Mathematics
• Keywords: Market Uncertainty ; Benford Distribution ; First Significant Digits ; Empirical Likelihood

Abstract

The distribution of first digits (FSD) in number series obtained from different data sets of natural and socio-economic processes show certain patterns. These patterns are marked by an asymmetry in favor of small digits. In this work we use empirical likelihood methods to estimate the First Significant Digit Distribution (FSDD) of the Credit Default Swap (CDS) market during the financial crisis 2007-2009. The resulting FSDD and Mean Preserving Spreads are used to model price uncertainty on this largest credit derivative market. During the last financial crisis we identify two periods of high uncertainty: One at the beginning of the crisis (Sep. 2007) and one at the time around the Lehman default. These periods can be interpreted in the way, that market participants were not able to identify the price generating distribution.