Whether or not the trend is your friend can depend on how well you identify it. Get on the right side of the trend and you might do very well. Get it wrong and you could be in trouble. Indicators are tools that can help you find the trend. There are a number of indicators, so one key question is: how do you know which one to use?
A problem traders face in determining which indicator to use is that trends are easy to see when they have already happened, but may be difficult to predict for the future. Using the backtesting capability of Fidelity Wealth-Lab Pro,® we can measure what would have happened if we had bought a stock after each of the trend-following indicators identified a trend whose direction is up. Comparing the results is a practical way to see if a trend lasts after the indicator spots it.
While backtesting can provide historical comparisons, those comparisons are not guarantees of future results. Still, these results can prove instructive if you are unsure which indicators you might like to use in the future.
A very important point to note is that indicators do not usually comprise comprehensive trading strategies; instead they may be used as building blocks to create a strategy of your own. The trend-following indicator may be combined with other techniques to more closely pinpoint trade setups, and fundamental analysis may also help to identify particular stocks to trade as well as their relative value.
A trend is just a series of price points moving in a particular direction over time. In an uptrend, it can be as short as a short-term price increase that lasts a few trading sessions, or it could run much longer, lasting for years. Indicators may try to measure the rate of increase, the volume associated with a price increase, prices compared to historic rolling averages, and more.
Here are the indicators we studied, a brief description of what they are, and the settings being tested.
|Average Directional Movement Index (ADX)||Uses the DX measurement, which looks at smoothed directional lines to measure the strength of price movement.||Buy with ADX above 20 and rising when the DI+ crossed above the DI- to signal uptrend.
Sell with ADX above 20 and falling when the DI+ crossed below the DI- to indicate a downtrend.
|Commodity Channel Index (CCI)||Tries to identify cyclical turns comparing the mean price of a security and the average of means over a period.||Buy when CCI broke 100.
Sell when CCI fell back below 100.
|Channel Break-Outs (NH/NL)||Creates a buy signal when the price rises above its 100-day high.
||Buy when close exceeded the high of the past 100 days.
Sell when price closed at a new 100-day low.
|Moving Averages (MA)||Compares price to a moving average of price—which is average price over a set time period.
||3 variations tested:
Buy price moving above average.
Sell price falling below average.
Buy when faster moving average crosses up through slower moving average.
Sell when faster moving average crosses down through slower moving average.
Buy moving average trending up.
Sell moving average trending down.
|Moving Average Convergence/ Divergence (MACD)||Converts the moving averages of short and long periods of time into a single line indicating if the lines are approaching one another or spreading.
The line itself is then averaged, and compared to the current position. Buy signals may be created when the signal line crosses zero, when the histogram and the signal line diverge, or when the price and the MACD signal line diverge.
|Buy when MACD signal line crosses average.
Sell when MACD line crosses moving downwards.
|MACD Histogram (MACDH)
||Difference between the MACD line and its signal line.
||Buy when MACDH is below zero and increasing.
Sell when MACDH is above zero and decreasing.
|Momentum (Mo)||Measures the difference between the current price and the price a specified period earlier as an absolute price difference. Positive values show an uptrend, negative values show a downtrend.||Buy when 100-day momentum is positive.
Sell when 100-day momentum is negative.
|Rate of Change (RoC)||Measures the percentage difference in price over the specified period. Like momentum, positive values show an uptrend, negative values show a downtrend.||Buy when 100-day momentum is positive.
Sell when 100-day momentum is negative.
|Linear Regression (LinReg)||Looks at the prices for the number of specified periods and finds a straight line which best fits all the prices. Use the slope to determine if the trend is up (positive value) or down (negative value), as well as the general strength of the trend.||Buy when slope is positive.
Sell when slope is negative.
|Chaikin Money Flow (CMF)||Measured the volume weighted average of accumulation/distribution over a specified period. The principle behind the Chaikin Money Flow, is the nearer the close is to the high, the more accumulation has taken place. Conversely the nearer the close is to the low, the more distribution has taken place. If the price action consistently closes above the bar's midpoint on increasing volume then the Chaikin Money Flow will be positive. The indicator has an oversold zone, and a buy signal is generated when price action develops a lower low into oversold zones and the CMF diverges with a higher low and begins to rise.||Bought when CMF crossed above zero.
Sold when CMF dropped below zero.
How we tested the indicators
Trade size: In the simulated scenario, each trade was executed using $10,000 (though the results are scalable for smaller amounts of capital).
Stocks: Current S&P; 500® index components (note this suffers from survivorship as extreme losers are eliminated from the index).
Stops: 10% below the buy price.
Out of more than a dozen tests, only four indicators returned more than a buy-and-hold strategy (see chart below). While the results aren’t shown in the table below, we tested these strategies over 200 and 50 days, but 100 days using these indicators proved to be the best performer. 100 days may line up with the cycle of the market during this time period (a six-month cycle) where short-term blips look to have been weeded out and we were able to focus on longer-term trends.
|Indicator||RoC100||MA100||Mo100||Linear Reg 100||Buy & Hold|
|Number of Trades||12,625||12,625||12,625||4,743||500|
|Average Profit %||1.80%||1.80%||1.79%||4.69%||37.64%|
|Average Bars Held||27.62||27.62||27.62||71.54||1,244.30|
|Results are for illustration only. All results in this article are based on testing all the stocks in the S&P; 500 Index as of 5/24/11 over a five-year period ending on that date. The results do not reflect commissions, fees, or taxes, which could reduce the returns. The results are according to Wealth-Lab Pro.(R) For definitions of these and other investing terms, please visit the Fidelity glossary.|
Little distinguishes the winning indicators from each other. In technical terms, the Linear Regression had a slight edge in win rate while the MA100, RoC100 and Mo100 topped the net profits. It appears that the deciding factor between them is a matter of preference.
In general, the more complex indicators did not perform as well as simpler rate-of-change indicators. See table below.
|Buy & Hold||$1,893,263.92||500||37.64%||1,244.30||66.00%||$2,466,593.58||($573,329.66)||($3,484,399.20)|
|Results are for illustration only. All results in this article are based on testing all the stocks in the S&P500; Index as of 5/24/11 over a five-year period ending on that date. The results do not reflect commissions, fees, or taxes, which could reduce the returns. The results are according to Wealth-Lab Pro.(R) For definitions of these and other investing terms, please visit the Fidelity glossary.|
For all indicators tested, the number of trades goes up dramatically versus Buy and Hold, and not only because of the trend changes (note the impact that trading costs would have on a more active strategy, compared with the buy and hold strategy). The trades got stopped out and re-entered almost immediately when the downdraft was not enough to upset the prevailing trend.
A key advantage to each of the indicators was spending a small fraction of the time in the trade, giving the potential to find other ways to profit when the capital is not in use.
The Average Bars in Trade was measured in days for all indicators. Even the winning trades, on average, did not spend more than a year in trade. The losing trades, being mostly stop losses, often exited in less than a month, on average.
The indicator win rates ranged from 24% to 48%, with the majority between 30% and 40%. Compare that to 66% for Buy and Hold. The MACD Lines Crossing turned in the highest indicator win rate at 49%, and that's significant because MACD's average profit per trade was less than 1%.
All trend-following indicators tested appeared to reduce risk by limiting drawdowns (a maximum loss less) considerably. The Moving Average Crossover had the least drawdown in this test set.
Many traders prefer to use these trend-followers in combination with other strategies. However, based on the results of this simulation, when you want to get aligned with the market trend, one helpful method might be to look at the current price relative to six months (100 trading days) ago. Those stocks that are up over the six-month period may have created an up-trend.
Past performance is no guarantee of future results.