Scenario generation for long term fuel prices

In this article we deal with the problem of scenario generation for fuel prices in the long term. The solution of many decision making problems in the energy sector such as the optimal mix of energy productions among different technologies, requires to model the dynamic of fuel prices and forecast their possible scenarios over time. We present two different approaches for scenario generation: a Vector autoregressive approach and a Monte Carlo approach; The first one is based on the estimate of a Vector Auto Regressive model i.e. a set of simultaneous equations. The second one is based of the assumption that the returns dynamics follow a generalised Weiner process. Using the two approaches we forecast prices’ scenarios.