The search for non-traditional models to predict the prices of financial markets still is a highly researched topic for an obvious reason. Unlike traditional methods, machine learning techniques allow to process a vast amount of information of different nature in order to extract non-random patterns hidden od price behavior in input data. Input data can be historical data on prices and volumes (high frequency included), other types of data as fundamental indicators, data on investor sentiment and attention provided by Google Search or using large language models. The objective of a potential dissertation thesis on this topic would be to create suitable machine learning models that are able to extract relevant information signals at the first stage and use them as inputs in a machine learning model of different setting to predict the future price as well as its movement direction of financial assets at the second stage. The results will be submitted for publication in journals like Expert Systems with Applications, Decision Support systems, International Journal of Forecasting.
References
- Henrique, B. M., Sobreiro, V. A., & Kimura, H. (2019). Literature review: Machine learning techniques applied to financial market prediction. Expert Systems with Applications, 124, 226-251.
- Ayala, J., García-Torres, M., Noguera, J. L. V., Gómez-Vela, F., & Divina, F. (2021). Technical analysis strategy optimization using a machine learning approach in stock market indices. Knowledge-Based Systems, 225, 107119.
- Zhong, C., Du, W., Xu, W., Huang, Q., Zhao, Y., & Wang, M. (2023). LSTM-ReGAT: A network-centric approach for cryptocurrency price trend prediction. Decision Support Systems, 169, 113955.
- Vuong, V. D., Nguyen, L. H., & Goulet, J. A. (2024). Coupling LSTM neural networks and state-space models through analytically tractable inference. International Journal of Forecasting. In Press.
