Multi-horizon ternary time series forecasting

Htike@Muhammad Yusof, Zaw Zaw (2013) Multi-horizon ternary time series forecasting. In: 17th Conference on Signal Processing Algorithms, Architectures, Arrangements, and Applications (SPA 2013, 26-28 Sep 2013, Poznań, POLAND .

	PDF (TOC) Restricted to Registered users only Download (186kB) | Request a copy
Preview	PDF (Proc Cover page) Download (345kB) | Preview
	PDF - Published Version Restricted to Registered users only Download (679kB) | Request a copy

Abstract

Time series forecasting techniques have been widely applied in domains such as weather forecasting, electric power demand forecasting, earthquake forecasting, and financial market forecasting. Because of the fact that these time series are affected by a multitude of interrelating macroscopic and microscopic variables, the underlying models that generate these time series are nonlinear and extremely complex. Therefore, it is computationally infeasible to develop full-scale models with the present computing technology. Therefore, researchers have resorted to smaller-scale models that require frequent recalibration. Despite advances in forecasting technology over the past few decades, there have not been algorithms that can consistently produce accurate forecasts with statistical significance. This is mainly because state-of-the-art forecasting algorithms essentially perform single-horizon forecasts and produce continuous numbers as outputs. This paper proposes a novel multi-horizon ternary forecasting algorithm that forecasts whether a time series is heading for an uptrend or downtrend, or going sideways. The proposed system utilizes a cascade of support vector machines, each of which is trained to forecast a specific horizon. Individual forecasts of these support vector machines are combined to form an extrapolated time series. A higher level forecasting system then forward-runs the extrapolated time series and then forecasts the future trend of the input time series in accordance with some volatility measure. Experiments have been carried out on some datasets. Over these datasets, this system achieves accuracy rates well above the baseline accuracy rate, implying statistical significance. The experimental results demonstrate the efficacy of our framework.

Item Type:	Conference or Workshop Item (Full Paper)
Additional Information:	6919/34350
Uncontrolled Keywords:	cascaded SVMs; time series forecasting; ternary forecasting; multi-horizon forecasting
Subjects:	Q Science > Q Science (General)
Kulliyyahs/Centres/Divisions/Institutes (Can select more than one option. Press CONTROL button):	Kulliyyah of Engineering > Department of Mechatronics Engineering
Depositing User:	Mr. Zaw Zaw Htike
Date Deposited:	23 Jan 2014 16:45
Last Modified:	01 Jun 2015 09:39
URI:	http://irep.iium.edu.my/id/eprint/34350

Actions (login required)

View Item