Our Weekly Premium Signals report now includes mean-reversion signals for SPY ETF. The strategy that generates the signals is not data-mined but based on our robust PSI5 algo, which is essentially a formula from a text on probability and statistics that describes fundamental behavior of price action.
Detailed weekly mean-reversion entry and exit signals for SPY ETF are included in Weekly Signals, which now feature signals for seven strategies after this latest addition.
Note that any timing strategies, especially of the mean-reversion type, that claim to beat buy and hold CAGR, especially during uptrends, are usually over-fitted or artifacts of selection bias due to post hoc choice of a group of securities. Some even involve look-ahead bias (not as uncommon as some may think.)
The objective of a sound timing strategy that is not data-mined, manually or automatically, is to generate higher risk-adjusted returns, not higher absolute returns. In general, this is accomplished by generating lower drawdown at acceptable return levels. Leverage can be used if the risk profile allows it for generating higher absolute returns but this is not the primary objective.
MRSPYW portfolio backtest parameters
Market: SPY ETF
Time-frame: Weekly (adjusted data)
Strategy type: Mean-reversion, long-only
Algo: Based on PSI5
Backtest period: 01/29/1993 – 07/05/2018
Commission per share: $0.01
Trade entry: Open of next weekly bar after entry signal (no look-ahead bias)
Trade exit: Open of next weekly bar after exit signal (no look-ahead bias)
In the results below we compare the performance of the weekly strategy to buy and hold total return.
|Parameter||MRSPYW||SPY Buy and hold|
It may be seen from the above performance table that the mean-reversion strategy generates higher risk-adjusted returns than the buy and hold strategy: Sharpe for the strategy is 0.95 versus 0.57 for the moving average cross in SPY.
Below are the equity curve and underwater equity curve for the weekly strategy. Click on images to enlarge.
The strategy avoids taking any positions during bear markets. For 2008, the loss is only -2.4% versus 36.8% for buy and hold.
Below is a table of monthly returns.
It may be seen that the strategy stayed out of the market during the 2001 and 2008 bear markets due to application of a downtrend detection filter.
Below is a Monte Carlo simulation curve.
The Monte Carlo simulation indicates that there is less than 5% probability of a drawdown larger than 25%.
Weekly mean-reversion entry and exit signals for SPY ETF are included in Weekly Signals.
Disclaimer: No part of the analysis in this blog constitutes a trade recommendation. The past performance of any trading system or methodology is not necessarily indicative of future results. Read the full disclaimer here.
All charts were created with Amibroker
CFTC RULE 4.41 – HYPOTHETICAL OR SIMULATED PERFORMANCE RESULTS HAVE CERTAIN LIMITATIONS. UNLIKE AN ACTUAL PERFORMANCE RECORD, SIMULATED RESULTS DO NOT REPRESENT ACTUAL TRADING. ALSO, SINCE THE TRADES HAVE NOT BEEN EXECUTED, THE RESULTS MAY HAVE UNDER-OR-OVER COMPENSATED FOR THE IMPACT, IF ANY, OF CERTAIN MARKET FACTORS, SUCH AS LACK OF LIQUIDITY. SIMULATED TRADING PROGRAMS IN GENERAL ARE ALSO SUBJECT TO THE FACT THAT THEY ARE DESIGNED WITH THE BENEFIT OF HINDSIGHT. NO REPRESENTATION IS BEING MADE THAT ANY ACCOUNT WILL OR IS LIKELY TO ACHIEVE PROFIT OR LOSSES SIMILAR TO THOSE SHOWN.