Mean Reversion Strategies with Polynomial Regression Channels
The Black Book of Day Trading Strategies
1,000 complete strategies · 31 chapters · Full trade plans
Mean reversion is a effective and widely used concept in quantitative finance. The basic idea is that asset prices tend to revert to their long-term mean. Polynomial Regression Channels (PRC) provide a sophisticated and adaptive way to identify the mean and to trade on deviations from it.
The Mean Reversion Hypothesis
The mean reversion hypothesis, in the context of PRC, states that the price of an asset will tend to return to the polynomial regression line after deviating from it. The upper and lower channel lines provide a measure of how far the price has deviated from the mean. When the price touches or breaches one of the channel lines, it is considered to be at an extreme, and a reversion to the mean is expected.
Entry and Exit Rules
A simple mean reversion strategy using PRC can be defined as follows:
- Entry:
- Go short when the price crosses above the upper channel line.
- Go long when the price crosses below the lower channel line.
- Exit:
- Close the position when the price returns to the polynomial regression line.
Z-Score Formula for Entry:
A more quantitative approach is to use a Z-score to measure the deviation from the mean:
Z = rac{P(t) - \mu(t)}{\sigma(t)}
Where (P(t)) is the current price, (\mu(t)) is the value of the regression line at time (t), and (\sigma(t)) is the standard deviation of the residuals. A trader might enter a short position when (Z > 2) and a long position when (Z < -2).
Stop-Loss Placement and Position Sizing
As with any trading strategy, risk management is paramount. A stop-loss order should be placed to limit the potential loss on a trade. A common approach is to place the stop-loss at a multiple of the channel width above the entry price for a short position, or below the entry price for a long position. Position sizing should be based on the volatility of the asset and the trader's risk tolerance.
| Trade | Entry Price | Stop-Loss | Target Price | Outcome |
|---|---|---|---|---|
| Short | 105.2 | 106.0 | 103.5 | Win |
| Long | 98.5 | 97.7 | 100.2 | Win |
| Short | 104.8 | 105.6 | 103.1 | Loss |
Trade Example:
A quantitative trader notices that the price of a currency pair has moved three standard deviations above its 50-period, degree-2 polynomial regression line. The trader initiates a short position, with a stop-loss placed at four standard deviations and a take-profit order at the regression line. The position size is calculated to risk no more than 1% of the trading account.
Conclusion
Mean reversion strategies based on Polynomial Regression Channels can be highly effective, but they require careful implementation and risk management. The flexibility of the PRC allows it to adapt to changing market conditions, providing a more robust measure of the mean than traditional, static indicators. The next article will explore a different type of trading strategy: breakout and momentum strategies.