
The Shape of Sovereign Credit Risk
Macro fundamentals predict sovereign CDS spreads across countries but not across maturities, and the gap is informative about functional form rather than about the fundamentals themselves.
Academic research on statistical learning, with an eye on what works in practice.
Live Last read · last published · next check

Macro fundamentals predict sovereign CDS spreads across countries but not across maturities, and the gap is informative about functional form rather than about the fundamentals themselves.

A Bayesian neural network gives every portfolio weight a credible interval, and the interval says when not to trade.

It looks like Bayesian uncertainty. In finance, that resemblance is the trap.

Why Bayesian model averaging is just another way to pull coefficients toward zero — and why that matters for prediction.

What Happens When You Let Irrelevant Predictors Vary Over Time

Why Mispricing Deserves a Place in Your Crypto Factor Model

This is Out-of-Sample, a newsletter about Economics, Finance, Digital Assets.