Recursive Estimation of Dynamic Models Using Cook's Distance, With Application to Wind Energy Forecast

Abstract

This article proposes a new class of adaptive forgetting factors for the recursive estimation of time-varying models. The proposed procedure is based on Cook's distance of the new observation. It is proven that the proposed procedure encompasses the adaptive features of classic adaptive forgetting factors and, thus has a larger adaptability than its competitors. The proposed forgetting factor is applied to wind energy forecast, demonstrating advantages with respect to alternative procedures.