Those expensive, complex models for predicting market swings are a total waste of money 94% of the time.
March 24, 2026
Original Paper
When Does Volatility Model Selection Matter? Entropy Diagnostics and Pre-Registered Evidence Across 1,496 Assets and Eleven Asset Classes
SSRN · 6366718
The Takeaway
The study found that a simple moving average—calculable in seconds—performs just as well as high-end predictive models for the vast majority of assets. This could eliminate 86% of the computational costs for financial firms without increasing risk.
From the abstract
<div> <p>We study when volatility model selection has value, rather than which model wins on average. A single entropy statistic, computable in seconds from a trailing return window, identifies 94% of assets where a simple EWMA baseline suffices—eliminating 86% of model-fitting cost. A 2×2 in-sample attribution separates two mechanisms: filtered historical simulation (FHS) fixes unconditional VaR coverage, while model diversity reduces violation clustering. Walk-forward backtesting across 1,491