| Summary: | When modelling financial data, it is important to be able to capture when anomalies happen. Being able to forecast that a certain stock price will plummet or rise beyond the normal range of fluctuations is important for risk management, portfolio management and options trading. Since forecasting can be treated as a supervised learning method, it is important for us to carefully select the features.
In this report, we will explore various machine learning and statistical methods to accurately forecast the occurrence of financial extreme events. Additionally, we will discuss a method to distil features into the important ones.
|
|---|