- CRB PriceCharts - Studies

Stochastic

The PriceCharts Stochastic Formula calculates momentum based on the rate of change in the daily high, low, and closing prices. PriceCharts uses the following 9-day stochastic formulas:

Raw Stochastic Value = 100[(C-L9) / (H9-L9)], where C is the latest close, L9 is the lowest low for the last nine days, and H9 is the highest high for the same nine days.

Two smooth 3-day moving averages are plotted, %K and %D, according to the following formulas:

%K = 2/3 previous %K = 1/3 Raw Stochastic Value; %D = 2/3 previous %D = 1/3 new %K.

The K line is the solid line, while the slower D line is a dashed line. The major signals to watch for are divergence between the D line and price, and K line crosses the D line.

Values above 70 are considered over-bought while values below 30 are considered over-sold.

For a detailed explanation refer to: Technical Analysis of the Futures Markets by John J. Murphy.

Tips on Technicals: RSI vs. Stochastics

Bullish Consensus

The Bullish Consensus is a poll of leading analysts and advisors and is expressed as a percentage of bullish sentiment. This poll (conducted Since 1964) produces a daily value. The following are suggested guidelines for using the Bullish Consensus as a trading tool:

Trade with the trend of the Consensus. Look to enter the long side as the numbers rise, the short side as they fall. Trade contrary to the Consensus at extremes. Extreme Consensus levels vary from market to market, and historical data is available to help identify them.

The 50% area often constitutes support in a bull market and resistance in a bear market.

For a detailed explanation of the Bullish Consensus, contact:
Market Vane Corporation, PO Box 90490, Pasadena, CA 91109
(818) 395-7436 www.marketvane.net

Relative Strength Index

Step 1. The R.S.I. is determined by adding the net change of the up days and down days respectively, taken from the last nine (9) market sessions. Each total is then divided by nine (the number of days used).

Step 2. The up average is the divided by the down average, and to this answer, add one (1). Dividing 100 by the resulting number will produce a figure which is then subtracted from 100. The answer is the Relative Strength Index.

Step 3. To update the R.S.I. the following day, BOTH the up and down averages are multiplied by eight (8), then the day's up or down amount (if any) is added to the respective averages, dividing by nine (9) for the new averages, then proceed with Step 2.

Values above 70 are considered over-bought while values below 30 are considered over-sold.

For detailed R.S.I. information and explanation, refer to: New Concepts in Technical Trading Systems by J. Welles Wilder Jr., available from Trader's Library.

Tips on Technicals: RSI vs. Stochastics

Directional Movement Index (ADX)

Although the average directional movement index (ADX) isn’t used as frequently as some of the popular technical indicators, the ADX line has definite advantages because it filters out a lot of the false oscillator signals which are frequently given early in a move.

A longer-term trader can stay with trending positions longer by following the simple guidelines for the ADX line. According to research by computer trading expert Bruce Babcock (1731 Howe Ave., Suite 149, Sacramento, CA 95825), a climb by the ADX line above 40 followed by a downturn signals an imminent end to the current trend (whether up or down). When this signal is given, traders should take profits on existing positions. More aggressive traders can use this signal to consider taking positions for a possible move in the opposite direction.

The following charts show how the ADX works. The ADX line on the feeder cattle chart gave two signals during the year. The first downturn accurately marked the top in February, and the second downturn above the 40 level signaled a bottom in late summer. Note that the signal in late July was actually more than a month ahead of the actual bottom in September. The ADX warns you of an end to the trend. In this case, it gave you more than a month’s warning.

Like the feeder cattle signals, crude oil’s ADX gave two signals during the year, one at the summer low and the second at the winter high. Both signals were given by climbing above 40 and turning down.

The ADX signals by feeder cattle and crude oil signaled the end of one trend and the beginning of a new trend. But the ADX is not designed to signal a trend reversal. It only signals the end of the existing trend. A good example of not signaling a trend reversal is in T-Bonds. The end of the strong spring rally was accurately marked by the ADX signal in June. Then T-Bonds consolidated in a coil until the upside breakout in the fall. An ADX climb above 40 and downturn in November signaled another consolidation.

Usually, a commodity gives no more than a couple ADX signals during a year, unless the market has particularly volatile price action.

The ADX is less helpful during sideways markets. During extended consolidation periods, the ADX line will slip toward 10. When the ADX approaches 10, a major move is usually about to take place. But the ADX line doesn’t tell you which direction it will go. You have to rely on other indicators for the probable direction of the next move.

The ADX is part of the direction movement system introduced by J. Welles Wilder in his book, New Concepts in Technical Trading Systems. Wilder introduced a 14-day ADX, and Babcock has not found any good reason to vary this time period. We are plotting a 14-day ADX line in CRB PriceCharts.

We also plot the 7-Bar ADX on weekly charts. This indicator is calculated and read the same as the 14-Day ADX, but it’s calculated on a seven-week time period instead of 14 days. It provides a longer-term view of average directional movement.
Remember, on weekly charts, we switch to the next contract when the nearby contract expires. In some markets, this will cause some short-term distortions on indicators which use prices in the formula.

In summary, if the market is trending (whether up or down), the ADX line should be rising. During an extended consolidation period, the ADX line will slip toward a low number.

Calculating the ADX:

Here’s how to calculate the ADX to update your charts. In the discussion that follows, the following conventions are used:

N = number of days (we use 14)
TH = today’s high
TL = today’s low
TS = today’s settle
YH = yesterday’s high
YL = yesterday’s low
YS = yesterday’s settle
DM+ = directional movement up
DM- = directional movement down
ADM+ = average directional movement up
ADM- = average directional movement down
TR = true range
DI+ = directional indicator up
DI- = directional indicator down
DX = directional movement
ADX = average directional movement

The calculations are as follows:

1. Calculate DM+ and DM-:

a. If YH > = TH, and YL < = TL (i.e. if today’s trading is totally within yesterday’s range), then DM+ = 0; DM- = 0

b. If TH - YH = YL - TL (i.e. if the differences between highs and lows are the same), then DM+ = 0; DM- = 0

c. If TH - YH > YL - TL (i.e. if the differences between highs and lows are the same), then DM+ = TH - YH; DM- = 0

d. Otherwise (i.e. the difference between lows is more than the difference between highs), DM+ = 0; DM- = YL - TL

2. Calculate ADM+ and ADM-:

a. ADM+ = ((yesterday’s ADM+) * (N - 1)/N) + (DM+ * 1/N)

b. ADM- = ((yesterday’s ADM-) * (N - 1)/N) - (DM- * 1/N)

3. Calculate TR and ATR:

a. If YS < TL, TL = YS

b. If YS > TH, TH = YS

c. TR = TH - TL

d. ATR = ((yesterday’s ATR) * (N-1)/N) + (TR * 1/N)

4. Calculate DI+ and DI-:

a. DI+ = ADM+/ATR

b. DI- = ADM-/ATR

5. Calculate DX and ADX:

a. DX = (DI+ - DI-)/(DI+ + DI-) * 100 (i.e. the difference of DIs over the sum of DIs, always positive)

b. ADX = ((yesterday’s ADX) * (N - 1)/N) + (DX*1/N)

For detailed ADX information and explanation, refer to: New Concepts in Technical Trading Systems by J. Welles Wilder Jr., available from Trader's Library.

Tips on Technicals: The Directional Movement System

Williams %R

Williams %R is an overbought/oversold indicator developed by Larry Williams. In his original work, Williams examined ten trading days to determine the trading range, then calculated where today’s closing price fell within that range.

To calculate the %R, you must first detemine the highest high and lowest low for the length of the interval - this is the trading range for the specified interval. Once those values are determined, the general formula for the % R is as follows:

%R_t = ( (High_n-Close_t) / (High_n - Low_n) ) X 100

%R_t = the percent of the range for the current period

High_n = the highest price during the past n trading periods

Close_t = the closing price for the current period

Low_n = the lowest price during the past n trading periods

n = the length of the interval (we use 10 days)

Assume the market is Treasury Bills. The high for the past ten trading intervals is 9275 and the low is 9125. The closing price in the current period is 9267. If you substitute those values in the equation, you get:

%R = ( (9275-9267) / (9275-9125) ) X 100
= (8 / 150) X 100
= 5.33

Trading signals from Williams %R

This study attempts to use the market’s normal strength and weakness to forecast turns. Since it measures the current price relative to the trading range of the past 10-day period, it always attempts to sell strength and buy weakness.

The system attempts to measure overbought and oversold market conditions. The %R always falls between a value of -100 and 0. The trading rules are simple: When %R reaches -10% or lower, you sell; when it reaches -90% or higher, you buy.

Note that these values are reversed from normal thinking. The %R works best in trending markets, either bull or bear trends. Likewise, it is not uncommon for divergence to occur between the %R and the market.

As with all overbought/oversold indicators, it is often best to wait for the market’s price to change direction before placing your trade. For example, if Williams %R is showing an overbought condition (10% or lower), it’s wise to wait until the price turns down before selling that market.

It is not unusual for an overbought/oversold indicator to remain in an overbought/oversold condition for a long time period as the market’s price continues to climb/fall. Selling or buying simply because the market appears overbought or oversold may take you out of a trade long before the trend actually changes.

Commitments of Traders

• Commercial Positions: the difference between the Commercial Positions-Long (All) and the Commercial Positions-Short (All).
• NonCommercial Positions: the difference between the NonCommercial Positions-Long (All) and the NonCommercial Positions-Short (All).
• NonReportable Positions: the difference between the NonReportable Positions-Long (All) and the NonReportable Positions-Short (All).

Positive values represent Net Long positions while negative values represent Net Short positions.

CRB Yearbook Articles:

•  1996: Analyzing the Commitments of Traders Report
•  1990: Analyzing the Commitments of Traders Reports
•  1985: Analysis of the CFTC Commitments of Traders Reports Can Help You Forecast Prices
•  1980: Commitments of Traders Reports as an Aid to Price Forecasting

Commodity Futures Trading Commission (CFTC)

Historical Open Interest

The dashed line running parallel with the solid open interest line in PriceCharts is a six-year average of the total open interest.

CRB Yearbook Articles:

•  1997: Seasonality of Volume and Open Interest
•  1991: Seasonal Volume and Open Interest Analysis

Williams Accumulation/Distribution Index (AD)

The Williams Accumulation/Distribution Index (AD) was developed by Larry Williams. It attempts to measure market pressures, specifically looking for market divergence, and serves to measure market strength and sentiment. You should watch for instances of substantial divergence between the AD and the underlying price direction as a key to future price direction.

If a market continues to move into new high ground, for instance, the AD should follow suit. When the market makes several new highs but the AD fails to make new highs, it’s a warning that the market may be about to reverse direction. Conversely, if the AD fails to make lower lows while market prices drift lower, it’s a signal that the market may be about to move higher. In either case, divergence implies a reversal in the dominant trend may be near.

Once you spot divergence, initiate a market position when a clear break occurs in the trendline of the AD. This minimizes the possibility of taking a position before the actual trend reverses.

The AD is computed several different ways. Some computations normalized the index, while others added extra smoothing factors through the use of moving averages. Commodity Price Charts uses the following computations to create and chart the AD index.

As a starting point, Commodity Price Charts sets the initial value of the AD index to zero. From there, we perform the following comparisons, which are logically and mutually exclusive. Only one of the three can be valid to correctly measure the market’s accumulation or distribution.

The first comparison checks for accumulation, i.e., is current close higher than previous close? If the market is accumulating, then compute the difference between current close and low. Next, add that arithmetic difference to the Accumulation/Distribution Index. The procedure is:

If Close_t > Close_t-1,
then AD_t = AD_t-1 + (Close _t – Low_t)

The second comparison checks for no change in price. If correct, the AD index does not change. It states:

If Close_t = Close_t-1, then AD_t = AD_t-1

The last comparison checks for a down market. It checks for current close below previous close. If that is correct, the market is distributing. First compute the difference between current high and close. Then subtract that difference from the AD index. This measures market distribution. The final computation is:

If Close_t < Close_t-1,
then AD_t = AD_t-1 - (High_t - Close_t)

AD_t = the accumulation/distribution index for the current period

AD_t-1 = the accumulation/distribution index for the previous period

Close_t = the closing price for the current interval

Close_t-1 = the closing price for the previous interval

High_t = the true high price for the current interval

High_t-1 = the true low price for the current interval

NOTE: The “true high” and “true low” are used in these formulas. The true high is the higher value of the current high or the previous close. The true low is the lower value of the current low or the previous close.

For detailed ADX information and explanation, refer to: New Concepts in Technical Trading Systems by J. Welles Wilder Jr., available from Trader's Library.

Tips on Technicals: The Directional Movement System