Hybrid wavelet-neural network models for time series

• Published in: Applied soft computing 2023-09, Vol.144, p.110469, Article 110469
• Main Authors: Kılıç, Deniz Kenan, Uğur, Ömür
• Format: Article
• Language: English
• Subjects: Long short-term memory (LSTM); Multiresolution analysis (MRA); Nonlinear models; Recurrent neural network (RNN); Time series analysis; Wavelet neural network (WNN); Wavenet
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2023.110469

The use of wavelet analysis contributes to better modeling for financial time series in the sense of both frequency and time. In this study, S&P500 and NASDAQ data are separated into several components utilizing multiresolution analysis (MRA). Subsequently, using an appropriate neural network structure, each component is modeled. In addition, wavelets are used as an activation function in long short-term memory (LSTM) networks to form a hybrid model. The hybrid model is merged with MRA as a proposed method in this paper. Four distinct strategies are employed: LSTM, LSTM+MRA, hybrid LSTM-Wavenet, and hybrid LSTM-Wavenet+MRA. Results show that the use of MRA and wavelets as an activation function together reduces the error the most. [Display omitted] •The study bridges the gap between hybrid models using MRA and hybrid models with WNN.•MRA improves the results of LSTM for financial time series.•Wavenet structure decreases the errors in neural networks.•Combining MRA and wavenet provides the best improvement in the analysis.•Changing dilation and translation parameters in each WNN node is insignificant.