Lagrange polynomial and cubic spline interpolations as the alternative procedures for estimating smoothing parameter in the single exponential smoothing method

Abstract

In this paper, we propose the alternative methods to estimate the smoothing parameter which is used on the Exponential Smoothing Methods. The present study is focused on estimating one smoothing parameter in the Single Exponential Smoothing (SES) Forecasting Method. This research provides an algorithm to estimate the smoothing parameter utilizing the 3^rd order Lagrange polynomial interpolation and cubic spline interpolation. Furthermore, this study is provided by an example of the use of these methods. The results from these proposed alternative methods for estimating the optimal parameter are then compared to the results obtained from the Excel solver. This research shows that the estimation result of the smoothing parameter in SES with cubic spline interpolation is able to produce a better estimation than the 3^rd order Lagrange polynomial interpolation and Excel solver.