Summary of An Evaluation Of Deep Learning Models For Stock Market Trend Prediction, by Gonzalo Lopez Gil et al.
An Evaluation of Deep Learning Models for Stock Market Trend Prediction
by Gonzalo Lopez Gil, Paul Duhamel-Sebline, Andrew McCarren
First submitted to arxiv on: 22 Aug 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This study investigates the efficacy of five advanced deep learning models in predicting short-term trends in stock markets. The models explored include Temporal Convolutional Networks (TCN), Neural Basis Expansion Analysis for Time Series Forecasting (N-BEATS), Temporal Fusion Transformers (TFT), Neural Hierarchical Interpolation for Time Series Forecasting (N-HiTS), and Time-series Dense Encoder (TiDE). Additionally, the study introduces the Extended Long Short-Term Memory for Time Series (xLSTM-TS) model, an xLSTM adaptation optimised for time series prediction. The models were tested using daily and hourly closing prices from the S&P 500 index and the Brazilian ETF EWZ. Wavelet denoising techniques were applied to smooth the signal and reduce minor fluctuations, providing cleaner data as input for all approaches. Denoising significantly improved performance in predicting stock price direction. Among the models tested, xLSTM-TS consistently outperformed others, achieving a test accuracy of 72.82% and an F1 score of 73.16% on the EWZ daily dataset. This research provides valuable insights into the application of machine learning for market movement forecasting, highlighting both the potential and the challenges involved. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The study aims to predict short-term stock market trends using advanced deep learning models. The researchers tested five different models, including TCN, N-BEATS, TFT, N-HiTS, and TiDE, as well as a new model called xLSTM-TS. They used data from the S&P 500 index and the Brazilian ETF EWZ to train the models. To make the data cleaner, they applied wavelet denoising techniques to smooth out minor fluctuations. This helped the models predict stock price direction more accurately. The best-performing model was xLSTM-TS, which achieved a high accuracy rate of 72.82% and an F1 score of 73.16%. This research shows that machine learning can be used to forecast market movements, but it also highlights some challenges involved. |
Keywords
» Artificial intelligence » Deep learning » Encoder » F1 score » Machine learning » Time series
