Ponente
Descripción
Advances in computing power and solution techniques have made stochastic optimization via scenario analysis a popular method for solving complex planning problems, including those with integer variables. This approach allows uncertainty to be considered as an additional element, resulting in valid solutions for representative set of scenarios. However, classical models are risk neutral and do not consider the possibility of catastrophic losses in extremely unfavorable situations. Therefore, alternative approaches have emerged that incorporate risk measures, such as Conditional Value- at-Risk, a consistent Var-based measure. This talk will present methods for incorporating risk aversion into natural resource management problems, such as forestry and mining. We will discuss alternatives that allow risk management throughout the planning horizon, rather than only the last period.
Key word: stochastic optimization, scenario analysis, measure of risk, Conditional Value-at -Risk
