Already have an account? log in
This article is FREE for you
Create a free account and join institutional investors, analysts and strategists from the world's largest banks
By submitting your email address you agree to receive communication by email
Good News! The Algos Say The US Economy ‘Is Doing Just Fine’
Listen, there's a lot of debate going on right now about late-cycle dynamics.
Chief among market pa
OR, subscribe now for unlimited access
By submitting your email address you agree to receive communication by email
By submitting your email address you agree to receive communication by email
It might be BAML’s first attempt but some of us have been succeeding with it for a long time. And the correct LP space weighted metric makes a big difference in executing “nearest neighbors”. Also, I’ll bet BAML is ignoring some of the most important variables….just like in 2007.
FWIW, not every algo is predicting such a long gap before the next recession: https://intensity.com/resources/intensity-recession-forecast-may-1-2018
With such a tiny dataset, kNN is probably going to do as well as any other technique, by which I mean shitty and worthless for trying to extrapolate through a discontinuity.