Stochastic optimization applied to financial portfolio management

This project covers the basics of Financial Portfolio Management theory through different stochastic optimization models. The concepts of uncertainty and stochastic models are introduced and proven to be more reliable than the deterministic models. The input data of the model now are random variables, and two types of stochastic models are analyzed in this project. The first type are the models that transform the random input into deterministic data before the model is run. Among them there are the classic Markowitz mean-variance (which is shown in a practical example for IBEX 35) and Black-Litterman models. The second type are the models that introduce the uncertainty of the input data explicitly in the models, which are prepared to handle with random variables. These models are advanced and complex to solve, but a detailed introduction to them is provided in this thesis: key features, general formulation and a multistage stochastic optimization modelling example with the same case of IBEX 35