Most traders are familiar with a set of market-timing indicators that are described as oscillators. Their goal is to identify trading opportunities which may exist whenever the oscillator indicates a market is overbought or oversold.* Those who use oscillators typically relate market price activity to oscillator behavior over compa
* For those of you unfamiliar with the terms overbought and oversold, essentially an oversold oscillator reading will be recorded when selling has persisted over an extended period of time and price has been declining, and an overbought oscillator reading will be recorded when buying has persisted over an extended period of time and price has been advancing.
rable time periods. The relationship between successive price and successive indicator movement over the same time period is important to some traders who rely upon divergence analysis. They believe that if the two movements diverge for any period of time, this disparity will be resolved in favor of the indicator. In other words, the movement of the oscillator should precede and therefore predict price movement. For example, if over a period of time, price has recorded two successively lower price lows and, over the identical period of time, the oscillator has failed to record a succession of two lower lows, then a positive divergence exists and the expectation of these traders is for price to rally. Conversely, if price has recorded two successively higher highs and the oscillator fails to confirm a succession of two higher highs, then a negative divergence exists and the expected outcome is for price to decline.
Despite the widespread acceptance of divergence analysis among traders, we are convinced that the tendency for price activity and oscillator activity to diverge at possible market turning points is merely a coincidence and to attribute it to divergence analysis is incorrect. We believe divergence analysis has many shortcomings: (1) a series of more than two successive price/oscillator divergences can often occur and consequently produce a series of premature, false price reversal signals; (2) a series of price/oscillator divergences can occur over an extended period of time thus reducing the effectiveness of this tool; and (3) a single overbought or oversold oscillator reading can occur without any secondary movement into this oversold/overbought zone, thereby preventing any divergence analysis.
We believe duration analysis, or the amount of time in which an oscillator resides in overbought or oversold indicator zones, is a more important determinant of market conditions and price reversals than divergence analysis. Our research indicates if the period of overbought or oversold lasts six or more price bars, then it is considered to be a severe or extreme reading. Severe readings imply that the market still has a great deal of momentum left and traders best not expect a price reversal at that time. In fact, if a trader were inclined to do anything, it should be to participate in the prevailing flow of the market. On the other hand, if the overbought or oversold period is limited to five or fewer price bars, then the duration is described as a mild reading. Generally, whenever a mild recording exists, the market experiences a price reversal. If an extreme oscillator reading is recorded, however, then a trader must await a subsequent mild reading to indicate a likely price reversal and justify entry. We feel this is the reason many traders who apply divergence analysis have some degree of success—usually an extreme reading is first recorded and is then followed by a mild secondary reading. It's more than pure coincidence but it has to do with frequency, or amount of time spent in overbought or oversold, rather than divergence.
Although the preceding comments are applicable to almost all market-timing oscillators, we are partial to using them with the ones we have developed. Many other widely followed oscillators are exponentially calculated and, consequently, have a tendency to create faulty readings. Also, the calculations for these oscillators require closing price comparisons from one price bar to the next. Because of this requirement, unscheduled market closings due to electrical failures or bad weather or unexpected news or political events such as assassinations or earthquakes can all skew closing price levels, thereby distorting the overall oscillator reading, not only for that trading session but for future sessions as well. Due to the inherent problems associated with these conventional exponential indicators, the oscillators we have developed are arithmetically calculated which have the advantage of removing any aberrant price data. In addition, we prefer to compare price levels other than consecutive price bar closes to remove the distortions caused by news which may affect consecutive price bar comparisons. For example, our TD REI indicator compares the highs and lows of alternating price bars.
TD RANGE EXPANSION INDEX (TD REI)
TD REI is a mathematically derived oscillator. Calculating TD REI manually is a tedious process since it requires the identification of reference prices, price comparisons, and qualifiers, as well as a series of mathematical steps. Fortunately, with the advent of computers this task has become simplified, and a long list of securities can be calculated and reviewed in a matter of minutes. Therefore, if you intend to use this oscillator and you are prone to arithmetic miscalculations or want to review many securities at one time, we suggest that you program the formula and rely upon the computer output for your results. The computation that follows is intended to give you an appreciation for the logic involved and the series of steps required to calculate a reading for TD REI. This way, TD REI will not simply be a black-box formula where practitioners blindly plug values in and obtain results. The five steps necessary to arrive at a (five-period) TD REI oscillator are as follows:
Step One. Rather than comparing the current price bar to the previous price bar as is done in calculating most oscillators, step one in the calculation of TD REI requires that one compare the current price bar to the price bar two periods earlier. More specifically, the high of the current price bar is compared to the high two price bars earlier; similarly, the low of the current price bar is compared to the low two price bars earlier. By performing this alternating price bar comparison rather than a successive price bar comparison, the impact of news-driven price distortions is reduced and the appearance of a much smoother oscillator read ing is apparent. It also increases the likelihood that a discernible price trend has been established. In this first step, one must subtract both the current bar's high from the high two bars earlier and the current bar's low from the low two bars earlier. These differences are then added together to create a summed value. Keep in mind that this summation can be either positive or negative, depending upon whether the current bar's high or low is greater than or less than the respective high or low two price bars earlier.
Step Two. Step two determines whether step one should be performed or a zero should be assigned to the oscillator reading. It does so by applying comparison qualifiers. The rule states that either the current price bar's high must be greater than or equal to the low five or six price bars earlier and the current price bar's low must be less than or equal to the high five or six price bars earlier, or the high two price bars earlier must be greater than or equal to the close seven or eight price bars earlier and the low two price bars earlier must be less than or equal to the close seven or eight price bars earlier. If neither condition exists, then a value of zero is assigned to the oscillator sum for that price bar, and the high-to-high and the low-to-low price bar comparisons described in step one are ignored. This adjustment reduces the likelihood that TD REI will record a premature overbought or oversold reading when markets are advancing or declining both rapidly and steeply.
Step Three. Step three adds the positive and the negative values from step one (if any) and the zero values from step two (if any) for each of five consecutive price bars and that value becomes the numerator. Next, the absolute value of the summed differences from step one (meaning the summation, whether positive or negative, is treated as a positive value for each of the most recent five price bars) becomes the denominator. In other words, the numerator can be positive, negative, or zero, and the denominator can only be positive. Once the numerator is divided by the denominator, a positive or a negative ratio appears. This value is then multiplied by 100 to create a percentage.
Step Four. Step four establishes an indicator band. Usually, a five-period TD REI fluctuates between -45 and +45, with the former level defining oversold and the latter defining overbought.
Step Five. Step five requires an analysis of price behavior once the TD REI has recorded an oversold or an overbought reading. In order to record a mild oscillator reading, TD REI must not spend more than five consecutive price bars in either the overbought or the oversold zone. Coincident and subsequent to a mild oscillator reading, oscillator conditions are ripe for a potential price reversal provided that a trigger mechanism, such as TD POQ (see the following section), generates and confirms a low-risk entry. Recording more than five consecutive price b^rs in the oversold or the overbought zone creates a severe or an extreme oscillator reading which suggests that the market's momentum is sufficiently intense to perpetuate its current movement, either upside or downside. Until the oscillator moves back into the neutral zone and then records a subsequent mild oversold or overbought reading, the prevailing market trend should continue. A countertrend entry may be indicated once a mild oscillator reading is recorded, provided TD POQ confirms, but it would be canceled if the duration of this move consumes more than five consecutive price bars.
Figure 8.1 of the 30-minute Johnson & Johnson chart demonstrates those situations where the TD REI has recorded extreme overbought and extreme oversold conditions. These readings are marked with a 6 next to the TD REI oscillator, which indicates that the market has recorded at least six price bars in the overbought or the oversold region. We mentioned that these extremities were not determined by divergences, but by duration. In the cases where the market for Johnson & Johnson was extremely oversold, and the TD REI has remained below -45 for a period of more than five price bars, one can see how it was best to avoid purchasing the security or a call option at that time, as further downward price movement occurred. Similarly, in the case where the market is extremely overbought, and the TD REI has remained above 45 for a period of more than five price bars, one can observe how it was advantageous to avoid selling the security or purchasing a put option at that time, as further upward price movement occurred.
TD PRICE OSCILLATOR QUALIFIER (TD POQ)
Most traders view an oversold market as a low-risk buying (call-buying) opportunity and an overbought market as a low-risk selling (put-buying) opportunity. As we described previously, they fail to make the distinction between degrees of overbought or oversold. By overlooking the critical difference between mild and severe or extreme overbought and oversold readings, they are operating at a distinct trading disadvantage. Even after they acquire the knowledge to distinguish between the two types of oscillator readings, however, traders must apply a device or trigger mechanism which signals that the price prerequisites have been fulfilled to confirm a low-risk entry opportunity. We rely upon TD Price Oscillator Qualifier (TD POQ) to perform this function for us.
Once TD REI has established an overbought or an oversold environment, TD POQ must be applied to the coincident price action in order to receive entry confirmation. This indicator is critical to the qualification and disqualification of an
FIGURE 8.1 If an oscillator remains in overbought or oversold for an extended period of time, it implies that the indicator reading is severe or extreme. Trend reversals usually coincide with mild overbought or oversold readings. In the case of extreme readings, a trader would be wise to defer anticipating a reversal.
Was this article helpful?