Please use this identifier to cite or link to this item:
Title: Poisson autoregressive models with exogenous variables and their application to Asian handicap football betting
Authors: Xerri, Julia (2018)
Keywords: Soccer -- Germany
Soccer -- Betting -- Germany
Regression analysis
Time-series analysis
Issue Date: 2018
Citation: Xerri, J. (2018). Poisson autoregressive models with exogenous variables and their application to Asian handicap football betting (Bachelor's dissertation).
Abstract: We study a class of Poisson autoregressive models with exogenous variables (PARX) that can be used to model and forecast a time series of counts. We look at the properties of the models, including conditions for stationarity and existence of moments. The reduced form of the PARX models is also considered, resulting in time series Poisson generalised linear models (GLM) with the identity link. Estimation techniques for both the general and reduced forms are studied. In particular, we apply the PARX models in their GLM confi guration, for predicting the outcomes of football matches. For this purpose, we look at two football teams of different ability in the German Bundesliga: FC Bayern Munich and Borussia Mönchengladbach. We use two seasons starting from August 2013 as our training set, and two seasons starting from August 2015 as our test set. Results show that the autoregressive aspect and the exogenous input both contribute in capturing the dynamics of the number of goals scored in a football match. We also see that PARX models coupled up with the Skellam distribution allow us to derive a suitable model for the distribution of the goal differences in a football match. Betting strategies applied to the Asian handicap market, informed by goal difference forecasts, are used to investigate the predictability and profitability of the mentioned betting market. Hypothesis tests related to the efficiency of this betting market are also included. Findings result in substantial winnings for the Borussia Mönchengladbach matches, though not very high levels of profi tability are achieved for FC Bayern Munich.
Appears in Collections:Dissertations - FacSci - 2018
Dissertations - FacSciSOR - 2018

Files in This Item:
File Description SizeFormat 
  Restricted Access
2.88 MBAdobe PDFView/Open Request a copy

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.