Sydney Financial Mathematics Workshop

Sponsored by

National Australia Bank

Financial Modelling in the Horserace Betting Context: Theory and Practice

David Edelman
University of Wollongong

and

Carrot Dicing - A Model for Tote Price Distribution in horse racing

Charles Magri

and

Parimutuel Derivatives

Peter Cotton
Stanford and Morgan Stanley

Time: 5:15--7:15 pm
Date: Monday 2nd December 2002
Venue: Ground Floor, AAP Seminar Room, 259 George St

Abstracts

David Edelman:   Financial Modelling in the Horserace Betting Context: Theory and Practice

A general discussion of financial modelling methods for horserace betting is presented, including direct analogies between Technical and Fundamental Analysis in the more conventional financial markets. Models specific to racing will be discussed, including the traditional Multnomial Logit and a new Stratified Support-Vector Machine (Neural Network-type) model. Additionally, applications of research in this area to Information Theory will be touched upon.

Charles Magri:   Carrot Dicing - A Model for Tote Price Distribution in horse racing

A model is presented for the price distribution for horses as seen on the Tote ( NSWTAB ). The model is based on randomised allocation of funds to each horse, respecting the order of favouritism. Some analytical attempts using order statistics and transformation of variables are presented also. The model is supported by two pieces of evidence:

  1. The empirical distributions for the horses over a period of 8 months (Saturday meetings in the major metropolitan centres),
  2. A method of exploitation (strategy) using the model producing profitable results extending beyond an initial observation set of 3 months.
The model has potential for generalisations in other areas of finance which will be discussed as well as directions for future research.

Powerpoint presentation CarrotDicing.ppt OK with PowerPoint 2000, Mathemaitcs is garbled with StarOffice on Linux.

Peter Cotton:   Parimutuel Derivatives

I will talk about the connection between racetrack markets and financial markets, including some (I think) under-appreciated but fundamental features of derivatives which facilitate alternative trading mechanisms. I recently filed a U.S. Patent on methods for effecting multilateral derivative contracts, but I'm certainly not the lone ranger on this one. More or less direct adaptations of parimutuel markets have been taken up by DB, Goldman, Bear Stearns et al. I'd like to explain why this is the tip of the iceberg, rather than a novelty market for trading weather, payroll, labour statistics and the like.

Please feel free to bring this to the attention of interested colleagues.