Please use this identifier to cite or link to this item:
Title: Multi-step ahead forecasting using Cartesian Genetic Programming
Authors: Kalganova, T
Dzalbs, I
Keywords: Cartesian Genetic Programming;Forecasting method
Issue Date: 2017
Citation: Inspired by Nature - Essays Presented to Julian Miller on the Occasion of his 60th Birthday
Abstract: This paper describes a forecasting method that is suitable for long range predictions. Forecasts are made by a calculating machine of which inputs are the actual data and the outputs are the forecasted values. Furthermore, it is the Cartesian Genetic Programming (CGP) algorithm that finds the best performing machine out of huge abundance of candidates via evolutionary strategy. The algorithm can cope with non-stationary highly multivariate data series, and it can reveal hidden relationships among the input variables. Multiple experiments were devised by looking at several time series from different industries. Forecast results were analysed and compared using average Symmetric Mean Absolute Percentage Error (SMAPE) across all datasets. Overall, CGP achieved comparable to Support Vector Machine algorithm and performed better than Neural Networks.
Appears in Collections:Dept of Electronic and Computer Engineering Embargoed Research Papers

Files in This Item:
File Description SizeFormat 
FullText.pdfEmbargoed until 01 Jan 2030570.12 kBAdobe PDFView/Open

Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.