It is a fundamental principle of the Decision Analysis that the subjective forecasts used in na analysis should be based upon a synthesis of all the available evidence. Thus, when part of the decision-maker s evidence is in the form of a variety of forecasting models, or expert opinions. Decision Theory requires him to formulate a combination of these predictors. This work takes into account the Bayesian methodologies Outperformance and Quasi-Bayes, as well as the classical model of Optimal Combination, all applied to the combination of petroleum medium prices, generated by experts fo Petrobrás, for several international markets. It is presents a theoretical description of the methodologies followed by a comparative analysis between performances of individual forecasts and combinations, and these among themselves.