An Uncertainty Estimation Model for Algorithmic Trading Agent

We present an uncertainty estimation model for algorithmic trading agents in financial markets. These agents utilise deep learning to analyse market data and identify trading opportunities. Our model allows for optimised trade execution by providing estimated uncertainty scores along with market predictions from deep learning models. Our model is tested in trading with the S &P 500 index. Experiments demonstrate that the estimated uncertainty correlates to prediction accuracy and increased profitability can be obtained with our proposed method.