Dimensionality reduction techniques to support insider trading detection
By Adele Ravagnani, Fabrizio Lillo, Paola Deriu, Piero Mazzarisi, Francesca Medda, Antonio Russo
Rating
1383
Battle Count: 20
Relevance
8/10
Highly relevant for market surveillance and regulatory compliance in quantitative trading operations
Implementation Complexity
7/10
Requires implementation of PCA and autoencoders, as well as careful parameter tuning
Reproducibility
3/5
The method is well-described, but the data is not publicly available due to privacy concerns
About this paper
Methodology: Unsupervised anomaly detection. Problem types: Anomaly Detection, Dimensionality Reduction, Unsupervised Learning.
The interactive Everscope explorer (charts, battles, favorites) loads below.