Ongoing Research
Written Work
- Hassoun, Z., Powell, B., MacKay, N. 2025. Kairosis: Kairosis: A method for dynamical probability forecast aggregation informed by Bayesian change point detection (International Journal of Forecasting)
- Hassoun, Z. 2025. Bayesian characterization of prediction market trader beliefs using filtered Beta distributions (in prep)
- Hassoun, Z. 2025. Markov Decision Processes for dynamic strategy in FA Women’s Super League football (in prep)
Overview
My research focuses on forecasting and knowledge aggregation in markets, sports, and geopolitical domains. I develop Bayesian ensemble models that dynamically weight expert and algorithmic forecasts using latent regime structure, with applications to prediction markets, macro forecasting, and sports analytics. Current work includes Bayesian change-point models for forecast aggregation and posterior inference of beliefs in market microstructure data, as well as Markov Decision Processes to model sequential strategic decision-making in football.