A survey on uncertainty quantification in deep learning for financial time series prediction
Survey on uncertainty quantification in deep learning for financial time series prediction, analyzing models and techniques.
What it examines
This survey reviews uncertainty quantification methods in deep learning for financial time series prediction, focusing on Bayesian neural networks, Monte Carlo dropout, and other techniques. It aims to address the challenges of predicting financial assets like stocks, Forex, and cryptocurrencies by analyzing model types, financial inputs, and prediction spaces.
What it concludes
The study concludes that there is significant potential for future research in UQ for financial time series prediction, particularly in combining different analysis methods and exploring Forex markets. Potential applications include improved financial forecasting and risk management strategies.
Evidence objects
The study concludes that there is significant potential for future research in UQ for financial time series prediction, particularly in combining different analysis methods and exploring Forex markets. Potential applications include improved financial forecasting and risk management strategies.
key_findings bullet 1 · key_findings · validation V0
Raw abstract and provenance
Investors make decisions about buying and selling a financial asset based on available information. The traditional approach in Deep Learning when trying to predict the behavior of an asset is to take a price history, train a model, and forecast one single price in the near future. This is called the frequentist perspective. Uncertainty Quantification is an alternative in which models manage a probability distribution for prediction. It provides investors with more information than the traditional frequentist way, so they can consider the risk of making or not making a certain decision. We systematically reviewed the existing literature on Uncertainty Quantification methods in Deep Learning to predict the behavior of financial assets, such as foreign exchange, stock market, cryptocurrencies and others. The article discusses types of model, categories of financial assets, prediction characteristics and types of uncertainty. We found that, in general terms, references focus on price accuracy as a metric, although other metrics, such as trend accuracy, might be more appropriate. Very few authors analyze both epistemic and aleatoric uncertainty, and none analyze in depth how to decouple them. The time period analyzed includes the years 2001 to 2022.
Source row: 93 · abstract type: unknown