When I was writing fiction, I avoided coincidences. Readers don't believe in coincidences, I was taught and went on to teach others. It's the sign of a lazy writer.
Coincidences happen more in real life than they do in fiction, however, and one happened recently. Just after sending in a recent Trader's Corner article on the Granville's OBV or on-balance-volume indicator, I picked up an old STOCKS & COMMODITIES magazine. An article referenced OBV when introducing another indicator that links price and volume. Granville's OBV was created to "uncover hidden coils in an otherwise noneventful, nontrending market," Buff Pelz Dormeier wrote in "Between Price and Volume" (25: 7, 21-28).
Dormeier wasn't writing about OBV, however. Dormeier was introducing another indicator that links price and volume: the VPCI or volume price confirmation indicator. If you didn't read the prior article on OBV, you might be wondering, what's the big deal about indicators that link price and volume? The big deal is actually two big deals. The first is that volume tends to lead price.
The second is that price and volume are independent measurements. Many indicators rely on inputs related only to price. However, that means that the indicator itself correlates rather well to the price action. While such indicators can offer hints about the strength of the price action, they can't be considered independent of that price action.
Volume is different. As mentioned in the Trader's Corner article several weeks ago discussing OBV, volume and price are non-correlative. Prices can move up while volume moves down. Price can decline while volume rises. Volume can be huge and prices can stay where they are or can move in a matching huge range. As Dormeier states, volume measures "participation, enthusiasm and interest" (21). The two are independent of each other.
What's the problem with correlative indicators? In an article on neural networks, Connie Brown employed an example that pinpoints the problem. "As an example," Brown writes in "Neural Networks with Learning Disabilities" (STOCKS & COMMODITIES, V. 11:5, 207-214), "it is generally accepted . . . that Eurorates and the price of gold are significant factors to forecasting the US dollar against the major currencies." Trying to use such highly correlated inputs to forecast movements of the US dollar against the major currencies is an example of one "major trap of financial forecasting." While highly correlated inputs might confirm price movement, they may not be as helpful in forecasting it.
Dormeier cautions that OBV and VPCI are used differently. While OBV picks out "hidden coils" in disorganized markets, VCPI depicts the health of trends already underway. Dormeier believes that VCPI gives earlier signals and that backtesting show it would have provided superior gains on lower risk. (I offer caution about relying on backtesting results when making trading decisions. Test it for yourself in real-life trading.)
A portion of a TM weekly chart provided in the article shows how VPCI looks.
Annotated Weekly Chart of TM (Stocks & Commodities, V. 25:8, 26):
Someone must have agreed with Dormeier's conclusions about the usefulness of VPCI. Dormeier's work on the VPCI proved so masterful that he won the Market Technicians Association's 2007 Charles H. Dow Award for a work that either used established techniques in innovative ways or broke new territory. Consider this: in its press release noting the conferring of the award, the MTA said it grants the award only if it discovers a work that warrants it. It's not automatically given each year. At the time Dormeier received it, three years had passed since the MTA had last granted the award.
Unfortunately, a canvass of other technicians turned up information that VPCI didn't appear to be included as an available indicator on StockCharts, QuoteTracker or QCharts. Stocks & Commodities did provide formulas and programs for the VPCI for TradeStation, MetaStock and eSignal, as well as several other charting programs. Those subscribers who use those charting platforms could try the indicator.
Calculation of the VPCI requires several steps, with each step meant to determine whether VPCI is positive, confirming the trend, or negative, contradicting it, and to weight each confirmation or contradiction. After the VPCI is constructed, a smoothed VPCI can be added as was done on the chart shown above. Crossovers up through or down below the smoothed VPCI signal confirmation or contradiction of a trend.
Those of us using QCharts, Stockcharts or QuoteTracker do not yet have access to
the VPCI, but Dormeier's work still confirms the importance of at least
considering some non-correlative studies in our technical analysis. Until now, I
typically have preferred my volume patterns rather pure, studying price spreads
in conjunction with volume patterns, but then I don't tend to trade
directionally. For those who do and can employ the coding provided by STOCKS &
COMMODITIES, it might
be worthwhile to check out Dormeier's VCPI as a method of
confirming or contradicting trends. Those using QCharts, Stockcharts or
QuoteTracker still have OBV and other studies combining price and volume
studies, but Dormeier's work has still helped. It's pointed out the necessity of
understanding just how a given indicator is best used.