Pyramidal method of extrapolation for short time series. (5th December 2019)

• Record Type: Journal Article
• Authors: Turbal, Yuriy; Bomba, Andriy; Turbal, Mariana; Sokh, Anastasiia; Radoveniuk, Olena
• Is Part Of: International journal of computing science and mathematics. Volume 10:Number 6(2019)
• Page Start: 525
• Page End: 533
• Publication Date: 2019-12-05
• Subjects: time series -- extrapolation -- forecasting -- Newton-Gregory backward difference formula -- finite differences -- short time series -- cubic approximation -- pyramidal method; Mathematics -- Periodicals; Computer science -- Periodicals; Mathematics -- Data processing -- Periodicals; 510.285
• Languages: English
• ISSNs: 1752-5055

Abstract

The paper proposed a new method of short time series extrapolation, which can be used for predicting of economic, environmental and geophysical parameters. This method is based on the property of the rows of modified finite differences that the best cubic approximation is in the range of convexity. Numerical results show significant advantages of the proposed method in comparison with approaches to extrapolate, based on the use of polynomials, including Newton's extrapolation.