Our PSI5 mean-reversion strategy has recovered from March drawdown as it did in 2008 and is now in the black year-to-date. We also compare results to those of RSI2 strategy and include some general comments about mean-reversion.
Updated Monday, June 22, 2020. Performance updated as of close of Friday, June 19, 2020.
The PSI5 mean-reversion strategy is not data-mined but based on a formula from a text in probability theory that models price action dynamics.
PSI5 strategy in SPY ETF backtest details
Time-frame: Daily (adjusted data)
Strategy type: Long-only
Market: SPY ETF
Backtest period: 01/04/1993 – 06/19/2020
Commission per share: $0.01
Position size: Fully invested
Position entry and exit: Open of next bar
Current status: Flat since June 16, 2020
Equity curve with monthly returns and drawdown profile
The strategy recovered from March drawdown . Year-to-date return is +8.3%. Win rate in backtest period is 66.7% with a sample of 1038 trades. Profit factor is 1.56.
As a comparison, the RSI2 strategy with standard settings, long if RSI(2) < 10 and exit if RSI(2) > 70 was down -15.5% year on March 4, 2020. Year-to-date (June 19, 2020) the strategy is down -9.2%.
In fact, no settings combinations in increments of 10 for RSI(2) entry (10 to 40) and exit (50 to 90) have generated positive performance for RSI2 as of March 4, 2020.
The problem is of course you would not know which settings to use in advance but only in hindsight. For example, from 1993 to 2019 (30, 90) has generated best performance. But with these settings, the strategy would be down about 8.3% year-to-date. The second best, (40, 80), would have resulted in loss of 11.7% year-to-date. The reason for this is that RSI2 is not a model of price action dynamics, as it is the case with PSI5.
A few comments about mean-reversion strategies
Mean-reversion strategies are highly risky because the trades usually go against the short-term trend.
In addition, the risk of these strategies is high because stop-loss cannot be effectively used since it may destroy profitability in most cases. In a sense, mean-reversion strategies often profit from stop-losses of trend-following strategies.
Mean-reversion strategies are for professional traders that can manage risk and adjust it depending on market conditions.
Most amateur traders even when using a profitable mean-reversion strategy may lose money because of uncle point and fear to take risky but eventually profitable signals.
Charting and backtesting program: Amibroker
Data provider: Norgate Data
Technical and quantitative analysis of major stock indexes and 34 popular ETFs are included in our Weekly Premium Reports. Market signals for position traders are offered by our premium Market Signals service