Abstract
Overfitting is one of the major challenges encountered when using neural network models for financial time series forecasting tasks. The nature of financial time series data, often sparser than in other fields, tends to exacerbate the symptoms of overfitting. In this study, we propose a method to alleviate overfitting in neural network-based time series forecasting tasks through a novel data augmentation algorithm. The neural network structure employed in this research is a Long Short-Term Memory (LSTM) recurrent network. The proposed method gradually reduces the learning rate during the training process, whilst randomly augmenting the training data. To validate the performance of the proposed algorithm, we divided the VIX Index, into four sections and performed cross-validation. When compared to a standard LSTM network without the application of our algorithm, we confirmed that the Root Mean Square Error (RMSE) was reduced by 23%, 34%, 32%, and 34% in each cross-validation section respectively.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
