Temporal Aspects of Stock Price Prediction,

The financial market, characterized by its perpetual evolving nature, has long been an active research field. With the emergence of new technologies in artificial intelligence and the increasing availability of data, there is a growing need to uncover the relationship between historical data and actual predictions. Despite the abundance of studies in this area, few are the researches involving Bayesian Networks that are traditionally perceived as unsuitable for financial forecasting due to their inherent complexities. However, the S&P 500 index, known to be a good reflection of the status of the American economy, presents an ideal case study for this research. Comprising of a vast amount of companies across various sectors, the S&P 500 offers a rich dataset for analysis. Leveraging Partitioned Dynamic Bayesian Networks, originally developed for health data, provides a unique opportunity to test the evolving behaviour of the market. Their ability to change in structure provides a great advantage over the more traditional Dynamic Bayesian Networks. Great emphasis was dedicated on acquiring high-quality data tailored for this approach. Despite their complexities, the explainabililty that Bayesian networks inherent bring could in the future provide important information on hidden market dynamics.