Outliers - A Deep Dive (PART 2)

Dealing with extreme values as a quantitative researcher.

Introduction

---

This is the second part of our series on outliers. In the last article, we looked at outliers as our enemies and discussed ways to deal with them. Now, we will take the opposite approach and treat them as a potential source of alpha.

We will discuss techniques and ideas for finding alpha in the extreme values of our datasets. I’ve also added a lot of practical inputs and nuances I’ve found. Things you only get through experience. Understanding which effects (trend, risk premiums, reversals, divergent hedging pressures) are suited to different conditions and why we apply things differently based on our hypothesis.