The Effect of the Projection Time Frame on Projection Performance and Projection Performance Requirements

Abstract

As the projection time frame extends, it increases the uncertainty of future conditions of the energy industry development and compromises the performance of long-term projections. The assumption that the discrepancy between projected values and the actual data decreases as the projection year nears the reference year does not hold true in all cases. However, the general tendency toward narrowing the projection error range appears well-established. Estimates of the likely error for the projection variables facilitate making more well-grounded the choice of the acceptable level of complexity of employed economic and mathematical models. If an elaboration of these models leads to a change in the projection that is less than the acceptable projection error, the practicality of implementing such approaches may prove unwarranted. The fundamental possibility of narrowing down the uncertainty is implicated by objective patterns and trends; the system inertia inherent into the energy sector, and other factors. Freestyle task-specific constructing of hybrid “pipelines” of uncertainty-handling formalisms are capable of addressing any realistically formulated forecasting problems.

Sections 1.1 and 1.2 are co-authored with Svetlana V. Steklova.