Volatility-managed portfolio

Volatility targeting scales a position so that recent realized volatility stays near a constant. Moreira and Muir studied volatility-managed versions of the market and of common factors: take more exposure after quiet periods and less after turbulent ones. It is a risk overlay on an existing portfolio. It does not pick which asset is cheap.

Not investment advice. Nothing here is a recommendation to buy or sell anything. Backtests and historical studies are not guarantees, and a rule that worked in one sample can fail in the next. Trading can lose money. Links are for education, not endorsements. Do your own research.

Rules

Measure realized volatility over a short trailing window, such as a month of daily returns. Set the next period's weight proportional to a target volatility divided by that realized number, and cap the leverage. Apply the scale to a precommitted base portfolio, such as a stock index. Rebalance on the same calendar as the volatility estimate. Do not raise the target because last year's smaller target looked timid.

Where AI fits

AI can compute the trailing volatility, apply a leverage cap, and show the unmanaged portfolio beside the managed one under the same costs. It should not use future volatility. The interesting question is whether the overlay changes the path you can live with, not whether a paper's sample can be restated as a promise.

Markets

Liquid index futures or broad funds, and factor portfolios if you already hold them. The overlay needs a vehicle you can resize cheaply.

Timeframes

Monthly, or whenever the volatility window is refreshed. The window itself uses daily returns.

Risks

Cutting risk after a spike can miss the rebound that often follows. Leverage after a quiet period increases exposure just before the quiet ends. Volatility estimates are noisy. Caps and windows are easy to fit. This overlay can underperform the asset it manages for a long time.

← Back to all strategies