Hands-On Machine Learning for Algorithmic Trading

Legal and reputational risks

The use of alternative datasets may carry legal or reputational risk, in particular when they include the following items:

• Material Non-Public Information (MNPI) because it implies infringement of insider trading regulations
• Personally Identifiable Information (PII), primarily since the European Union has enacted the General Data Protection Regulation (GDPR)

Accordingly, legal and compliance requirements require thorough review. There could also be conflicts of interest when the provider of the data is also a market participant who is actively trading based on the dataset.