Premium Market Analysis

Trading Strategies

Mean-Reversion Strategy Performance

In this article we look at the performance of a mean-reversion strategy since January 2020. The backtest period includes the fast market crash in the first quarter of 2020 and offers a good test of strategy robustness.

Some long-only strategies in equity index ETFs take advantage of the mean-reverting property of the markets (buying the dips) to generate superior risk-adjusted returns. A good test of these strategies is the crash in the first quarter of 2020. This assumes that there was no adjustments of the strategy in hindsight.

In this article we look at the performance of PSI5 mean-reversion strategy from 01/02/2020 to 04/12/2021. The only parameter used by the strategy for the mean-reversion lookback period was not changed and it is the same as in previous articles.

The PSI5 mean-reversion strategy is not a data-mined but based on a formula from a text in probability theory that models price action.

PSI5 strategy 

Time-frame: Daily (adjusted data)
Strategy type: Long-only
Backtest period: 01/02/2020 – 04/12/2021
Commission per share: $0.01
Position size: Fully invested
Position entry and exit: Open of next bar

SPY ETF performance (backtest)

The strategy (green line) outperformed buy and hold (red line) during the 2020 crash and although the total return performance is close, risk-adjusted performance of the strategy is higher. Maximum drawdown for the strategy was 22% versus nearly 34% for SPY ETF. The strategy return was 20% in 2020 and it is 8.9% year-to-date.

QQQ ETF performance

The strategy (green line) outperformed buy and hold (red line) during the 2020 crash and for most part of the backtest. Maximum drawdown for the strategy was 18.7% versus 28.5% for QQQ ETF. The strategy return was 50.2% in 2020 and it is 16.7% year-to-date versus 46.1% in 2020 and 7.4% year-to-date for buy and hold.

Performance summary

SPY PSI5 SPY B&H QQQ PSI5 QQQ B&H
CAGR 30.7% 30% 75.2% 58.3%
MDD -22.2% -33.6% -18.7% -28.5%
SHARPE 0.96 0.74 2.06 1.27
TRADES 39 1 39 1
WIN% 74.4% 82.1%
EXPOSURE 34.8% 100% 36.1% 100%

From the above table it may be seen that PSI5 long-only mean-reversion strategy generated superior absolute and risk-adjusted returns as compared to buy and hold in the test period considered. For longer backtest results click here.


A few comments about mean-reversion strategies

Mean-reversion strategies are risky because by design the signals usually go against the short-term trend. Stop-loss risk management cannot be used because it usually destroys profitability.

Mean-reversion strategies are for professional traders who understand risk and can manage it well.

Amateur traders can lose money even when using a profitable mean-reversion strategy because of fear of taking signals that go against the short-term trend.

Usually professionals allocate a small percentage of equity to mean-reversion strategies.


Charting and backtesting program: Amibroker

Data provider: Norgate Data

More information about PSI strategy:

PSI mean reversion strategy information.

If you found this article interesting, you may follow this blog via RSS or Email, or in Twitter

Disclaimer