Time Series Analysis and Forecasting of In-Play Odds on a Betting Exchange

Bunyan, Andrew (2015) Time Series Analysis and Forecasting of In-Play Odds on a Betting Exchange. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (383kB) | Preview
Preview	PDF (Configuration File) Download (568kB) | Preview

Abstract

One of the newest phenomenon’s in the world of gambling, exchange betting has often been compared to the fluctuations and unpredictability of the financial markets. Within this unique technology is perhaps a more unambiguous form of gambling – in-play betting. This live betting scenario offers a rare insight into the peaks and troughs of betting odds throughout the duration of various sporting events. Horse racing is a sport unanimously associated with the punter and gambling and coupled with its in-play options, is an ideal area to analyse. In-play betting provides a plethora of potential analysis but perhaps the most intriguing of which is time series analysis. From start to finish, time stamped data of racing odds can be analysed to better understand the nature of the odds fluctuations and may even lay the foundations for potential race prediction and forecasting. This project will research horse racing data through time series analysis across a multitude of races whilst attempting to forecast a winning event (horse) based on the previous odds during the race, through the implementation of the R programming language and WEKA data mining interface.

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science G Geography. Anthropology. Recreation > GV Recreation Leisure > Games and Amusements > Gambling
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Caoimhe Ní Mhaicín
Date Deposited:	19 Oct 2015 11:21
Last Modified:	19 Oct 2015 12:23
URI:	https://norma.ncirl.ie/id/eprint/2098

Actions (login required)

View Item