The Journal of Prediction Markets

Moving Median Against Moving Average, Institutional Investors Against Private Investors and Market Efficiency

2025-07-09T19:37:09+01:00

A typical way of testing the hypothesis of market efficiency is to compare the performance of technical trading rules with the buy and hold strategy. In this work, an alternative technical rule is proposed based on moving median, and its predictive power for the Athens Stock Exchange (ASE) is contrasted with that of the most popular technical trading rule of moving average, with and without the true transaction costs existing in ASE. In the theoretical case of no transaction costs empirical findings show that the predictive power of the moving median rule is higher than that of the moving average rule, while both perform better than the passive strategy. Hence, for both trading rules the hypothesis of weak-form efficiency is rejected. By introducing true transaction costs and simulating various scenarios for the investors’ status, it is found that the moving median trading rule still outperforms the moving average one. Further, in terms of efficiency, evidence shows that it is still possible for an institutional investor to beat the specific market, even marginally, but this is not the case for a typical small investor, due to higher transaction costs for the latter. Consequently, the result on the testing of the hypothesis of efficient markets, given true transaction costs, depends on the status of the investor.

Artificial Neural Networks or Regression Modelling: Does it Matter?

2025-08-12T17:39:59+01:00

This study explores the ability of artificial neural networks (ANNs) to digest anomalies from factor models and investigates whether ANN models play a similar role in dissecting anomaly returns between developed and developing markets. The sample includes stocks from both the United States (U.S.) and Chinese markets, covering the period from 1995 to 2021. Neural network and traditional regression models are constructed using in-sample data, and their predictive performance is evaluated on out-of-sample data. Six well-known asset-pricing factors are selected as input variables, and the long-short spreads of nine anomaly strategies are the single output variable. The results show that the proposed ANN models outperform traditional regression models in holdout performance, regardless of the number of input factors or market type. Furthermore, ANN models in both the U.S. and Chinese markets demonstrate the same incremental accuracy performance. This provides evidence in support of using ANNs for financial modelling.

Psychological Predictors of Ethical Behavior in Investment Contexts: A Behavioral Forecasting Model

2025-08-21T10:16:11+01:00

This study investigates the predictive role of core psychological traits, risk tolerance, and demographic variables in shaping ethical investment behavior, with an emphasis on developing a behavioral forecasting model. Using data from 444 retail investors, the study examines the influence of Neuroticism, Self-Esteem, Self-Efficacy, and Locus of Control, the sub-traits of Core Self-Evaluation (CSE), alongside the risk tolerance and demographic factors. Multiple regression results reveal that among the CSE sub-traits, only Neuroticism significantly and negatively predicts ethical investment behavior. Risk tolerance and gender also emerge as significant predictors, whereas self-esteem, self-efficacy, and locus of control show no significant effects. These findings highlight the dominant role of emotional stability and risk disposition in forecasting ethically guided financial decisions. By integrating psychological dimensions into predictive modeling, the study contributes to the growing literature on behavioral decision-making and offers valuable insights for the design of ethically attuned investment strategies and prediction markets.

The Relationship Between Pre-Game Betting Volume and TV Ratings in NCAA Football

2025-10-17T13:06:25+01:00

The legalization of sports betting across the United States has transformed the way fans engage with collegiate athletics. This paper examines the empirical relationship between pre-game betting volume and national television ratings for NCAA football games, offering new evidence on how wagering and media consumption interact. Using game-level data from the 2024 season, we estimate two fixed-effects regression models—one predicting Nielsen household ratings and the other predicting pre-game betting volume—while controlling for game competitiveness, scoring potential, kickoff time, and broadcast network. Betting data were obtained from Action Network, while television ratings were collected from Sports Media Watch. Results indicate a strong, positive, and statistically significant relationship between betting volume and viewership: games with higher pre-game wagering activity attract larger national audiences, even after controlling for team and network effects. The absolute value of the point spread negatively affects both betting and viewership, consistent with lower engagement for less competitive contests. Broadcast network also significantly shapes outcomes, with games aired on ABC, FOX, ESPN, and CBS associated with greater betting volume and higher ratings. These findings suggest that betting markets and media audiences are mutually reinforcing, reflecting the role of wagering as both a consumption complement and a signal of fan interest. The study contributes to the growing literature on the convergence of sports media and gambling markets and highlights practical implications for networks, sportsbooks, and regulators in the post-PASPA era.