The Macro-OU chip keeps the same mean-reversion engine as the plain OU baseline but makes one honest upgrade: the equilibrium it reverts toward is allowed to move with the macro backdrop (credit stress, growth, liquidity), instead of being a single fixed number for all time. It is also the dependency-light fallback for the Macro-GBM model.
The plain OU model pulls sentiment back toward one unconditional long-run average. But 'normal' isn't the same in every environment — during credit stress or quantitative tightening, the index spends more time in fear, and that is the normal for those conditions. Macro-OU reflects this by sorting the macro backdrop into buckets and giving each bucket its own equilibrium, so stressed conditions revert toward their own historical fear level rather than toward a blended all-conditions mean. The reversion speed stays the same as the baseline; only the target shifts.
Macro-OU also serves as the dependency-light fallback for the more elaborate Macro-GBM model when the machinery Macro-GBM needs isn't available. As always, don't take the added macro-awareness as proof of a better forecast: the "Model performance" table under the Fear & Greed forecast shows Macro-OU's coverage, MAPE, RMSE, Brier and hit-rate beside the plain OU baseline, so you can see whether the market-conditions-dependent equilibrium actually helps on current data.
Same reversion engine and speed, but the equilibrium it reverts toward moves with the macro backdrop. Plain OU pulls toward one fixed long-run mean; Macro-OU pulls toward the level that is normal for the current market conditions.
Because during credit stress or tightening, sentiment genuinely spends more time in fear — that is the normal for that environment. Reverting toward a single all-conditions mean would fight the actual backdrop.
Not automatically — it adds macro-awareness, which can help or can overfit. The 'Model performance' table under the chart shows Macro-OU and OU side by side on coverage, MAPE, RMSE, Brier and hit-rate so you can compare them on current data.
Browse all S&P 500 tickers to see this metric applied to individual companies.
Educational research only — not investment advice.