Informing probability distributions in Equity Portfolio Optimization with Fundamental performances of stocks.

The aim of this work is to improve the results obtained in the literature regarding informing probability distributions in the field of financial markets. The main idea is related to Information Theory, as the goal is to inform the prior distribution with views or additional information external to the owned data, to obtain the posterior distribution modifying the prior in an optimal and meaningful view-based way. Specifically, being a Bayesian approach, the final goal for these frameworks is to inform historical data with experts’ or proprietary views, based on the concept of minimizing the divergence between the prior and the posterior distributions, namely, avoiding to introduce unknown information. Calculating the dissimilarity between probability distributions is not an easy task, but some divergence measures already exist and work pretty well in precise situations. My study in this direction is both to construct reliable views to inform historical data and to evaluate the existing models and divergence measures into the project and its conditions. Slightly modifying and adjusting the existing models, interesting new results are achieved.