This empirical and theoretical investigation into betting markets provides more evidence on the issue of the favourite-Iongshot bias, the nature of movers in betting markets and the impact of broken odds using a dataset which contains information on every race ofthe 2003 flat racing season. In the first chapter, the favourite-Iongshot bias is examined for different classes of horse races under the premiserlhat insider information (a cause for the existence of the bias) will playa larger role in lower class races. The evidence suggests that the bias exists for the 2003 data and that the odds quoted at the cessation of the market are more accurate than those quoted at the formation of the market. There is evidence of a stronger bias in the lower class races, but only for odds quoted at the formation of the market. The second chapter pursues an unanswered question posed by the first chapter which is also not flilly covered by the existing literature: how accurate are market moves? It is only known that the post-move odds are more accurate. The linear probability model from the first chapter is adapted to investigate the accuracy of market movers and yields the result that market moves are accurate on average. Finally, the issue of broken odds is examined as a possible cause of the favouritelongshot bias observed in the earlier chapters. The third chapter contributes to the literature by providing the first theoretical treatment on broken odds. The conditions derived suggest that for pari-mutuel markets the nature of the dividends returned due to breakage will cause returns to exhibit the phenomenon of a reverse favouritelongshot bias. For bookmaker markets, the nature of the quoted odds will cause a favourite-Iongshot bias, but not enough to account for all of the bias observed in the data.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:487310 |
| Date | January 2007 |
| Creators | Shing, Hui-Fai |
| Publisher | Royal Holloway, University of London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0188 seconds