Copy

☕️ ...

We expect the proof to be in the eating of the pudding.  But that does not mean it always is.

A lost decade in equity markets does not necessarily make us question the existence of the equity risk premium, so why does a year or two of bad strategy performance shake our conviction?

In this week's commentary, Nathan tackles this very topic.  He uses simulation techniques to demonstrate that even for investors with known accuracy rates, luck can create meaningful, phantom deviations over the short-term that can create over- and under-confidence for investors.

As it turns out, truly distinguishing between luck and skill in active management is a near-impossible task.

This is all confounded further when we realize that accuracy is only part of the equation: the magnitude of wins and losses can have a large impact as well.  So even though a strategy might feel like a winner, there might be a significant lurking loss out there (XIV, anyone?).  Conversely, a strategy might feel like a consistent loser, only to have a few, big wins that make up for the losses (e.g. trend following).

It's a quick read and a few graphs tell the whole story, so don't be afraid to dive in.  I think you'll walk away glad you did.

Read it on our blog or grab the PDF.

⚡️Corey


News

 👩‍💻HIRING: We're looking to hire for a Junior Quantitative Analyst role in Los Angeles, CA.  If you're interested or know someone who is, you can find the full job description here.

→ ✈️ TRAVEL: I'll be in the King of Prussia, PA area on March 20th.  If you'd like to get together, let me know!


Read of the Week
→ PREMIA: "The new-sample evidence reveals that the large majority of global factors are strongly present under conservative p-hacking perspectives, with limited out-of-sample decay of the premiums."  Global Factor Premiums


✉️ Join Newsletter | ✎ Send us your thoughts | ✄ Share URL

 

Copyright © 2019 Newfound Research LLC, All rights reserved.



Want to change how you receive these emails?
You can update your preferences or unsubscribe from this list.

Newfound Case #8291056