The Journal of Prediction Markets

A Comparative Study of Machine Learning Models For NCAA Men’s Basketball Tournament Games Outcome Prediction

2022-09-15T09:13:17+01:00

The evolution of machine learning has produced many avant-garde prediction algorithms, some of which provide more accurate predictions than conventional statistical tools. For the last two decades, machine learning (ML) approaches have largely been employed to predict game results. The purpose of this study is to utilize data gathered from web sources on the performance of 354 NCAA basketball teams over five seasons from 2015 to 2019 to forecast the results of NCAA men’s basketball tournament games with an assortment of big data and machine learning classification models. Among these include Sweet Sixteen, Elite Eight, and Final Four. The prediction results of each model were analysed and compared, and Decision Tree showed the best prediction performance compared with KNN, Logistic Regression, and SVM classification models, with a prediction accuracy of 75.71%, but Decision Tree was prone to overfitting problems. However, the Decision Tree is prone to overfitting problems, while Random Forest can correct the overfitting problem of Decision Tree by bagging and reduce the variance of Decision Tree prediction. Therefore, this study hypothesized that Random Forest would outperform Decision Tree in predicting NCAA game results. The results showed that, after a comprehensive
analysis and comparison of the evaluation metrics of the Decision Tree and Random Forest models, Random Forest was found to have better acceptable forecast performance than Decision Tree, with a prediction accuracy of 85.71%.

Models vs. Markets: Forecasting the 2020 U.S. Election

2022-09-21T15:17:42+01:00

We present a case study of forecasts for the 2020 U.S. presidential and congressional elections. Specifically, we investigate the accuracy of two competing forecast methods: model-based, statistical projections (from FiveThirtyEight.com) and prediction market prices (from PredictIt). We propose a market-based scoring (MBS) method for evaluating the performance of probabilistic forecasts. Our analysis finds that PredictIt and Five Thirty Eight perform comparably based on traditional metrics such as calibration and accuracy. For market-based scoring, however, we find that, if we ignore PredictIt’s fees and commissions, then Five Thirty Eight forecasts beat the markets overall, but if we factor in fees and commissions, the markets beat Five Thirty Eight. We discuss the implications of this analysis for forecasting future election cycles and betting market design and operations.

Comment: The first version of this article was posted on Researchers. One on October 26, 2020, 1 week before the 2020 U.S. election. That first version serves as a pre-registration of our proposed method before the election outcome is determined. All analysis performed below is based on the pre-registered methods proposed in the first version. The pre-registered article can be found at https://researchers.one/articles/20.10.00004.

Investing in March Madness: An Examination of The Relationship Between Sports Betting and Portfolio Construction

2022-10-19T00:30:38+01:00

This study focuses on three basic ideas: (1) sports betting should be considered an asset class; (2) sports betting contests, with two teams or contenders, are well described by a recently introduced probability distribution, the generalized Poisson binomial (GPB) distribution; and (3) Modern Portfolio Theory (MPT), Post-Modern Portfolio Theory (PMPT), and the Kelly criterion can be applied to yield optimal risk-return portfolios in such two contender environments. For the PMPT application, a unique quadratic-binary programming model is developed. March Madness data, based on the NCAA men’s annual championship basketball tournament, will provide examples of these portfolio theory approaches.

COVID-19 Pandemic and International Energy Performances

2022-10-21T23:45:46+01:00

The COVID-19 pandemic severely disrupted global capital markets and has continued to influence energy index returns since the beginning of 2020. Throughout this time, several key events relating to the pandemic have been observed to increase volatility, whilst others, in contrast, result in the opposite occurring. This study investigates the short-term volatility of these key events on global energy index returns. The data from six major energy indices were used to establish a global geographic perspective of energy market returns, specifically that of Asia Pacific, Australia, New Zealand, Europe, and the United States, including two commodities of WTI Crude Oil and Natural Gas to understand the effects of COVID-19 on energy returns. We employ “Event Study” by using a 10-day window period surrounding the dates of key events to measure volatility within returns. The findings of this study document that movement control orders increased volatility in energy market returns, whilst economic stability and vaccine availability tend to decrease volatility. The findings are crucial for investors, business owners, and government stakeholders to develop effective pandemic response plans whilst also providing insights on volatility expectations for investors to improve sentiment and confidence in navigating the stock markets under unpredictable conditions.

Economic Uncertainty and Bitcoin Volatility: Evidence During COVID-19

2023-01-09T10:05:46+00:00

This research investigates the predictability of economic uncertainty indexes on the volatility of Bitcoin (BTC) during COVID-19. The economic uncertainty indexes include US economic policy uncertainty (EPU), Twitter economic uncertainty (TEU), Twitter market uncertainty (TMU), geopolitical risk index (GPR), and trade policy uncertainty (TPU) index. The empirical findings show that the Twitter market uncertainty (TMU) and geopolitical risk (GPR) uncertainty index are valuable predictors of BTC volatility. Moreover, the combination forecasts information for all economic uncertainty indexes is useful for BTC volatility forecasting. Also, we find evidence during high and low volatility and the Russia–Ukraine war. Our results show that Twitter market uncertainty and geopolitical risk uncertainty index are effective predictors of Bitcoin volatility during high volatility periods. During the Russia–Ukraine war, economic policy uncertainty (EPU), the Twitter market uncertainty index, and combination forecast information for all uncertainty indexes are effective for Bitcoin volatility prediction. Our findings are robust with the alternative method MCS test.

The Wisdom of No Crowds: The Reaction of Betting Markets to Lockdown Soccer Games

2023-01-11T11:03:38+00:00

The support of home spectators is one of the contributing factors to the home advantage effect in sports matches. The COVID-19 pandemic led to European soccer matches being played without spectators. Contrary to previous findings in the literature, we show that betting markets adjusted promptly to account for a reduced home advantage in both goal difference and the probability of a win. These adjustments proved accurate over a large sample of soccer matches subsequently played without spectators, even though the earliest games appeared to suggest a much bigger change in home advantage.