LAUSR

This paper discusses the oil project optimal portfolio selection under uncertain environment where cash flows of the projects are mostly determined by experts’ estimations due to the lack of historical investment data. The oil project investment is usually distinguished by its high input, high risk and highly fluctuating ROI sensitive to the economic, political and technology uncertainties. Besides, in most of the cases, it is quite difficult to find reliable referential historical data for a specific project. All these peculiarities make actual oil project investment decision under high uncertainties. In this paper, we use normal uncertain variables to describe the cash flows and estimate the uncertainty distribution of the cash flows by experts’ experimental data. Then, under the constraint of controlling for bankruptcy, we give uncertain programming models to construct portfolios that maximize the expected returns and minimize the sine cross-entropy of the actual return from a prior return. Finally, we provide some numerical examples that fit different risk preference assumptions to further illustrate the feasibility and effectiveness of the models.