Wavelet-based forecasting of ARIMA time series - an empirical comparison of different methods
Journal Title: MANAGERIAL ECONOMICS - Year 2014, Vol 0, Issue
Abstract
By means of wavelet transform, an ARIMA time series can be split into different frequency com- ponents. In doing so, one is able to identify relevant patters within this time series, and there are different ways to utilize this feature to improve existing time series forecasting methods. However, despite a considerable amount of literature on the topic, there is hardly any work that compares the different wavelet-based methods with each other. In this paper, we try to close this gap. We test various wavelet-based methods on four data sets, each with its own character- istics. Eventually, we come to the conclusion that using wavelets does improve forecasting qual- ity, especially for time horizons longer than one-day-ahead. However, there is no single superior method: either wavelet-based denoising or wavelet-based time series decomposition is best. Performance depends on the data set as well as the forecasting time horizon.
Authors and Affiliations
Carola Deuschle, Stephan Schluter
Keywords
Related Articles
- EP ID EP167020
- DOI -
- Views 154
- Downloads 0
Negotiations as a tool of business strategy creation
The objective of the paper is to describe the author’s concept of considering the negotiations between internal stakeholders as a speci?c tool for the creation of company’s strategy. Firstly, the de?nition of such negoti...
Selected implications of negative externalities - on the example of the Polish energy sector
The aim of this paper is to present selected issues of the effective allocation of resources and its function in optimal decision-making processes. The authors focus on the economic consequences of market failure in the...
Wavelet-based forecasting of ARIMA time series - an empirical comparison of different methods
By means of wavelet transform, an ARIMA time series can be split into different frequency com- ponents. In doing so, one is able to identify relevant patters within this time series, and there are different ways to utili...
Integrated reporting - a "one-size-fits-all" solution!?
In a world of constantly growing streams of information, extensive transparency in public and private environment as well as recent ?nancial crises, it seems to be essential to enhance reporting to a higher, more sophist...
The relationship between WIG-subindexes: evidence from the Warsaw Stock Exchange
Evidence on links among the major Warsaw Stock Exchange subindexes using daily data from the period of 2011.03.01 to 2014.07.21 is provided in the paper. The empirical study for eleven WIG-sectoral subindexes with 844 ob...