From the June 01, 2011 issue of Futures Magazine • Subscribe!

# Trading stocks with stochastics

Technical analysis is a science of forecasting the price of tradeable markets based on price, volume and open interest data. Although those two variables — price and volume — imply a rather limited scope, hundreds of formulas and techniques have been developed to uncover their secrets.

These tools fall into two general categories: Pattern recognition and indicators. A chart pattern is formed within a chart when prices are graphed. Typically, patterns naturally occur and repeat. They are based on the visual interpretation of the data. Indicators, on the other hand, are less esoteric. They’re derived from mathematical formulas and are used to forecast a stock or commodity’s momentum, direction, etc. While the calculation is relatively fixed, the interpretation generally is open to opinion. As such, while they are hard-coded technical indicators, stochastics tend to be used in a discretionary way. (For a close look at stochastics systematic application in the gold market, as well as additional guidance on how to apply, and not apply, this indicator, see "Trading stochastics in context".)

Technical indicators fall into two broad categories: Those that are designed to lead the market and forecast price moves, and those that are designed to lag the market and facilitate post-move analysis. The relative strength index and stochastics are designed and interpreted as leading indicators. Examples of lagging indicators are moving averages or moving average-based indicators, such as moving average convergence-divergence. Typically, leading indicators work better in range-bound markets, while lagging indicators work best in trending markets.

Stochastics is a momentum indicator that signals a reversal in price momentum near key levels, ideally allowing you to enter the market with accuracy for bigger profits. George C. Lane developed the tool in the late 1950s.

Lane has said that the stochastic oscillator "doesn’t follow price, it doesn’t follow volume or anything like that. It follows the speed or the momentum of price. As a rule, the momentum changes direction before price." If that description is true, then stochastics should provide a valuable clue to future price direction.

The stochastic oscillator begins with a calculation of what percentage a security’s closing price is of its price range over a given time period, which is 14 periods (minutes, hours, days, weeks, etc.) by default. The formula to calculate that percentage, which is defined as %K in Lane’s terminology, is:

%K = 100[(C - L14)/(H14 - L14)]

Where:

C is the last traded price at closing.

L14 is the lowest traded price in previous 14 trading sessions.

H14 is the highest traded price in previous 14 trading sessions.

Next, a lesson from moving averages is applied. Moving averages are simply an average calculation over a set range. They are helpful for smoothing out noisy data streams and providing a reference for current data changes. As the range moves through time and each recent value is added, the oldest value is dropped. This way, the range remains the same. For stochastics, Lane calculated a three-period moving average of %K and plotted this on the same graph as %K. He called the three-period average %D.

The %D line also is called the trigger line. When the faster moving and noisier %K crosses above or below %D, it is considered significant, particularly when analyzed in the context of actual price action.

The stock market is rife with examples of stochastics correctly identifying market turns. Dell Inc. (see "Turn by turn") provides several.

The chart shows a long-term daily price chart of Dell Inc. The stochastics indicator uses the default setting of 14 days. The rectangle identifies an area where it worked particularly well to capture intermediate-term gains. The low of the period, on Dec. 8, 2009, was \$12.74. The high, when stochastics marked the end of the move on Dec. 28, was \$14.81.

Let’s use these data to demonstrate the calculation of the oscillator near the end of the move. The closing value on Dec. 28 was \$14.6. Putting this into the formula, we get:

%K = 100[(14.6 – 12.74)/(14.81 – 12.74)]

%K = 89.85

Low readings (below 20) indicate that price is near its low for the given time period. High readings (above 80) indicate that price is near its high for the given time period. Because our reading is well above 80, we could say that Dell Inc. was overbought at that price based on recent price action.

Fast, slow or full

The stochastic indicator can be categorized further into three types: Fast stochastic, slow stochastic or full stochastic.

The fast stochastic oscillator is based on original formulas for %K and %D. Fast stochastic plots the location of the current price in relation to the range of a certain number of prior bars (dependent upon user input, usually 14-periods). This is the indicator as it has been described in this article.

The slow stochastic oscillator simply adds another smoothing step to the original calculation. With the slow stochastic, %D becomes %K and yet another three-period moving average calculation is performed to generate a new, slower, %D:

Slow %K = Fast %D (three-period moving average of fast %K)

Slow %D = Three-period moving average of slow %K

The slow stochastic indicator is designed to create fewer false signals because of the additional smoothing effects of the moving averages.

The full stochastic indicator typically refers to a fully customizable version of the slow stochastic. Here, the analyst will modify the look-back period of the range, the number of periods used to tame the slow %K and, likewise, the number used in the calculation of the slow %D. Each version of the indicator is shown in "Three views".

Click the chart above to enlarge.

In the chart, %K is the blue line and %D is the red line. As we can see, the slow stochastic gives fewer false signals compared to fast stochastics. This is highlighted by the circles, which identify an area where false signals get smoothed out in the slower version of the indicator.

The stochastic oscillator can be used to generate buy and sell signals in a number of ways. First, traders can look for extreme value readings. These indicate overbought and oversold levels. Second, crossovers of the %K and %D lines are considered important. Third, divergences between the stochastics indicator and price tend to mark important areas of potential tops and bottoms.

Probably the easiest signal to identify is an extreme reading. The stochastic oscillator is plotted within a range of 0 and 100. It signals overbought conditions above 80. The indicator signals oversold conditions below 20. Most consider the move out of, not into, these ranges more important. In other words, someone following strict stochastics rules might buy when stochastics moves above the 20 line and sell when it moves below the 80 line, as a market can remain overbought or oversold for an extended period of time.

Note, however, that overbought readings are not necessarily bearish. Securities can become overbought and remain overbought during a strong uptrend. In the same fashion, oversold readings are not necessarily bullish. Securities also can become oversold and remain oversold during a strong downtrend (see "Apple rising"). In a trending market, it’s a good idea to ignore overbought and oversold readings.

Crossovers are another important signal for the stochastics analyst, particularly when paired with overbought and oversold levels. A trader using this strategy would buy when the %K line rises above the %D line while also moving up from below 20. Similarly, a sell would be signaled when the %K line fell below the %D line and did so while the indicator moved below 80 (see "Crossing lines"). Keep in mind, however, that crossovers are notorious for false signals, particularly when in the middle of the range.

Divergence occurs when the price trend and the indicator trend are moving in opposite directions. This signals that the direction of the price trend may be weakening as the underlying momentum is changing.

There are two types of divergence. Positive or bullish divergence occurs when the indicator is trending upward while the security is trending downward. Negative or bearish divergence gives a bearish signal as the underlying momentum is weakening during an uptrend. As with crossovers, divergences never should be taken without confirmation from either price or other indicators.

As seen in "Different directions," the leading Indian stock market index, Nifty Fifty, was making higher highs as indicated in the price chart, but stochastics was registering lower highs. After the market peaked around 6324, the Nifty started correcting, with the reversal being confirmed by a break of crucial support. Once the support was broken, the Nifty trended lower, correcting 8%.

Stochastics is popular for its ease of use, intuitive analysis and widespread distribution through popular trading software. It’s also a prolific indicator, offering an abundance of signals, through overbought/oversold readings, crossovers and divergence. As with all individual indicators, however, stochastics is best used as one view of the market. It should not be used in a vacuum; it is only one tool that can help your technical outlook. Traders should apply stochastics along with other indicators, such as trendlines, moving averages, Bollinger bands, volume and chart patterns (but not all of them).

The market, whether volatile, trending, choppy or smooth, can affect how well any indicator performs. If you know your market, and your indicator, you can put yourself in a good position to profit.

Bramesh Bhandari is an independent trader who trades the Indian stock market and teaches technical analysis at www.brameshtechanalysis.com.

Page 2 of 5