A trading strategy that combines volatility adaptive trend-following with mean-reversion to trade Dow 30 stock in daily timeframe outperforms SPY buy and hold total return since inception.
Edit: Backtests updated to October 23, 2020.
In another article we introduced a volatility adaptive moving average strategy that switches to mean-reversion when there is no trend and applied it to SPY ETF. The results exceeded expectations.
The long-only strategy in this article also employs the volatility adaptive moving average to identify and follow trends in stocks and switches to mean-reversion when there is no trend in the stock price. The parameters did not change except apart from the addition of a new rule.
For the backtests in this article we used Norgate data for Dow 30 index that also include current and past constituents to remove survivorship bias. We highly recommend this data service to those who would like to remove survivoship bias from backtests (we do not have a referral arrangement with the company.)
The strategy uses price action volatility to adjust the period of a moving average (AMAV) in a crossover that signals uptrend start and end in stocks. The adaptive moving average speed changes to better capture directional moves while minimizing losses due to whipsaw.
Furthermore, when the volatility adaptive moving average signals no trend in a stock, mean-reversion based on PSI5 algo is employed to maximize returns. The PSI5 mean-reversion strategy is not data-mined but based on a formula from a text in probability theory and is available for sale to professional traders and hedge funds subject to acceptance of a non-disclosure agreement.
Strategy general description (AMAVMRDOW):
Trend-following mode: AMAV > 200-day simple moving average
Mean-reversion mode: AMAV < 200-day simple moving average
Backtest details for AMAVMRDOW strategy
Time-frame: Daily (adjusted data)
Strategy type: Long-only
Market: Dow 30 constituents, past and current
Backtest period: 01/04/1993 – 10/23/2020
Commission per share: $0.005
Maximum open positions: 30
Position size: Equity/30
Position entry and exit: Open of next bar
Trend filter: S&P 500 above 200-day moving average (for AMAV part only)
Before including the results for AMAVMRDOW strategy, below are the results for just the trend-following strategy part.
Buy: AMAV > 200-day simple moving average
Sell: AMAV < 200-day simple moving average
The trend-following part of the strategy outperforms SPY total return buy and hold on a risk-adjusted basis but CAGR is lower by about 130 basis points. Maximum drawdown is about 20% for the strategy versus 55% for buy and hold.
In AMAVMRDOW strategy, mean-reversion is used during periods of tend-following inactivity in stocks. Below is the equity curve of the backtest.
|Parameter||Strategy||Buy and Hold (SPY TR)|
The strategy has outperformed SPY total return buy and hold on both absolute and risk-adjusted basis. CAGR for the strategy is 11.6% versus 9.8% for buy and hold. Sharpe for the strategy is 0.73 versus 0.52 for buy and hold.
The strategy backtest shows positive returns in 2008 bear market, in 2011 correction and +5% Year-to-date (October 23, 2020). The strategy may have potential assuming no major change in price action dynamics that will cause the PSI5 algo to not perform as expected and also cause degradation in adaptive moving average performance.
Also note that the above strategy has more than 10K trades and there is no survivorship bias because we used the stocks that were part of the index at the time of the signals. Most backtests in financial blogosphere show a few trades only and do not deal with the bias.
This AMAV code is offered as a bonus with a purchase of PSI5 mean-reversion algo.
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