Alternative methods of predicting competitive events: An application in horserace betting markets

Lessmann, Stefan; Sung, Ming‐Chien; Johnson, J.E.V.

doi:10.1016/j.ijforecast.2009.12.013

Cited by 33 publications

(24 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Practical use of the predictive models built in this study. [6], forecast air traffic delays [56], analyze the risk of mortgage prepayment [47], determine the likelihood that a customer will cease doing business with a company [75], predict horse race outcomes [77], and to evaluate the likelihood of being elected to the baseball hall of fame [33].…”

Section: Probabilistic Forecastmentioning

confidence: 99%

Application of machine learning to construction injury prediction

Tixier

Hallowell

Rajagopalan

et al. 2016

Automation in Construction

273

123

View full text Add to dashboard Cite

a b s t r a c tThe needs to ground construction safety-related decisions under uncertainty on knowledge extracted from objective, empirical data are pressing. Although construction research has considered machine learning (ML) for more than two decades, it had yet to be applied to safety concerns. We applied two state-of-the-art ML models, Random Forest (RF) and Stochastic Gradient Tree Boosting (SGTB), to a data set of carefully featured attributes and categorical safety outcomes, extracted from a large pool of textual construction injury reports via a highly accurate Natural Language Processing (NLP) tool developed by past research. The models can predict injury type, energy type, and body part with high skill (0.236 b RPSS b 0.436), outperforming the parametric models found in the literature. The high predictive skill reached suggests that injuries do not occur at random, and that therefore construction safety should be studied empirically and quantitatively rather than strictly being approached through the analysis of subjective data, expert opinion, and with a regulatory and managerial perspective. This opens the gate to a new research field, where construction safety is considered an empirically grounded quantitative science. Finally, the absence of predictive skill for the output variable injury severity suggests that unlike other safety outcomes, injury severity is mainly random, or that extra layers of predictive information should be used in making predictions, like the energy level in the environment. In the context of construction safety analysis, this study makes important strides in that the results provide reliable probabilistic forecasts of likely outcomes should an accident occur, and show great potential for integration with building information modeling and work packaging due to the binary and physical nature of the input variables. Such data-driven predictions had been absent from the field since its inception.

show abstract

Section: Probabilistic Forecastmentioning

confidence: 99%

Application of machine learning to construction injury prediction

Tixier

Hallowell

Rajagopalan

et al. 2016

Automation in Construction

273

123

View full text Add to dashboard Cite

show abstract

“…Interestingly, horseracing is the type of multicontestant sport that has attracted more attention, presumably due to its financial importance in betting markets. Lessman, Sung and Johnson (2010) have recently presented a model for predicting outcomes in horseracing, which can be used for predictions in all multi-contestant sports. The model they present is based on a much earlier work by McFadden (1974), which in turn contains propositions presented in Luce and Suppes (1965).…”

Section: Literature Reviewmentioning

confidence: 99%

“…In the classical work of McFadden (1974), the problem tackled is that of qualitative choice and it is coined as "conditional logit estimation." So, Lessman, Sung and Johnson (2010) see their problem as a problem of choice behavior: the eventual outcome of a multi-contestant race is the result of successive "choices." Starting from an initial set of n contestants, the winner is chosen, thus leaving n-1 contestants competing and successively reducing the number of contestants, until all contestants are ranked.…”

Section: Literature Reviewmentioning

confidence: 99%

Simulation of a Probabilistic Model for Multi-Contestant Races

Gakis¹,

Pardalos²,

Choi³

et al. 2018

AJSPO

View full text Add to dashboard Cite

Predictions of sports games have been recognized as an important area of study for their economic significance. Most models for such games cover two-player games and the resulting championships or study individual players or teams and their resulting comparative position. In this paper, we elaborate on a model for such multi-contestant races based on order statistics of the negative exponential distribution for race times. The games involve only a fixed number of athletes out of a broader pool of several athletes. Several large samples of games of ten athletes that play four at a time are generated, and Maximum Likelihood Estimators are calculated for the relative dominance parameters of the athletes. The estimated parameters are then used to test the predictive validity of the model. The results are discussed. Subsequently the model is modified for Erland-2 distributed race times. New samples are generated and the predictive validity of the new model is discussed. The paper concludes with a discussion of future research directions for improving the predictive validity and establishing confidence levels.

show abstract

“…The applications of RF have been growing because of their higher accuracy and their decreased sensitivity to noise in predicting binary targets, e.g., Jiang et al 13 in bioinformatics; Buckinx and Van den Poel, 8 and Prinzie and Van den Poel 10 in marketing; Larivi ere and Van den Poel 9 in economics; Chandra et al, 18 Bhattacharyya et al, 19 and Lessmanna et al 20 in finance; Coussement and Van den Poel 29 and Xie et al 30 in customer churn prediction; Culter et al 14 in ecology. Moreover, Evans and Cushman 15 provided good guidance for using RF; Falkowski et al 16 gave accurate demonstrations (>95% significance) of RF with lidar data; and Murphy et al 17 proposed model improvement ratio (MIR) for reasonable variable selection from RF variable-importance scores and overall model significance measurement for model fit.…”

Section: Background and Related Researchmentioning

confidence: 99%

Diagnosing Assets Impairment by Using Random Forests Model

Chen

2012

Int. J. Info. Tech. Dec. Mak.

View full text Add to dashboard Cite

This study develops a diagnosing model to examine the outcomes of assets write-off in enriching the literatures of assets impairment. Prior studies employed the Logit, linear and Tobit regression models to classify the determination of assets impairment and to diagnose the magnitude of the impairment, respectively. However, the drivers of assets write-off are somewhat complicated explicitly or implicitly, these models are unlikely to provide fairly satisfactory results. To improve the diagnosis, the Random Forests model is used for the classification determining and the magnitude diagnosing of assets impairment in this study. The result reveals that the Random Forests model outperforms the Logit and linear regression models in each case with variables selected by individual wrapping approach. This study also demonstrates diagnostic checks for both models with similar selected variables. The results are robust to these various specifications.

show abstract

Alternative methods of predicting competitive events: An application in horserace betting markets

Cited by 33 publications

References 23 publications

Application of machine learning to construction injury prediction

Application of machine learning to construction injury prediction

Simulation of a Probabilistic Model for Multi-Contestant Races

Diagnosing Assets Impairment by Using Random Forests Model

Contact Info

Product

Resources

About