*Dow Theory -- long-term trend method*

I conducted a quick search of the forums just a minute ago and was surprised to find only a handful of mentions of one of the oldest technical analysis schools of thought ever developed - Dow Theory.

For those not familiar with Dow Theory it eponymously named after the founder and first editor of the Wall Street Journal, Charles Dow. Dow is often considered the father of modern tehcnical analysis and was the first to create an index with the goal of measuring the overall movement of U.S. equities. Dow, who died in 1902, did not formally formulate what is known today as Dow Theory, however, several of his successors pieced together his letters and writings to build on Dow's discoveries of market timing to beat the averages. One of the best known experts on Dow Theory was Robert Rhea who in the '30s refined Dow's observations describing them in detail in his book The Dow Theory: An Explanation of Its Development and an Attempt to Define Its Usefulness as an Aid to Speculation.

So what is meant in the above numbered lists? First, the primary trend is inviolate. The primary trend is the longest of three trends Dow identified (primary, secondary, and minor trends). The primary trend is broad, long-term price movement and in colloquial conversation can be thought of as bull and bear markets lasting several years (typically two years or more). The secondary trend is and intermediate-term reaction against the primary trend that last typically from 3 weeks to 3 months or more and may often retrace 1/3 to 2/3 of the move from the last secondary trend's terminus (e.g. pivot high or pivot low). The minor trend is the day-to-day vacillations in the market. Dow believed the day to day movements and even the secondary trend could be manipulated but that the primary trend could not be manipulated - it was inviolate. And for this reason Dow thought the primary trend shold be the focus of serious investors.

The idea of confirmation is at the heart of the application of Dow Theory. Confirmation in Dow Theory occurs when both indexes reach a new secondary high or secondary low on a daily closing basis. The confirmation does not have to come on the same day and in reality may be several days or weeks apart. The primary factor to consider is that a trend cannot be deemed to have reversed unless both averages confirm a change in direction by setting new lows in the case of a primary bull market which is reversed or new highs in the case of a primary bear market. This is probably best demonstrated by examples which are included below.

With the evolution of the Dow Jones Industrial Average over the years to include quite a broad array of companies, many of which are not traditional goods manufacturers, and also the less representative nature of rails in the Transportation Index to include air, sea, and trucking companies there are other versions of Dow Theory floating around out there. One such popular methhod is to look at the S&P500 as representative of large-cap multinational companies many of which derive a good portion of revenue in non-dollar denominated markets and pair that with the small-cap index, the Russell 2000, which is primarily dollar-denominated domestic companies. We'll look at both below.

For the nitpicky the charts below are wrong. For the sake of demonstration I elected to use weekly data looking back three years. My opinion is the data looks much cleaner this way using the StockCharts platform, however, traditional Dow Theory looks at daily closing data and a line chart of daily closes would be sufficient. More importantly, what are the charts saying? It looks like in the case of the Industrials and the Transports the Transports are failing to confirm. This does not mean the primary trend has reversed. Recall that a new low must be put in by both averages. The failure to confirm is a significant event, particularly one that has persisted for over a year and it should be monitored, but it does not on its own reverse the primary trend. We are seeing a similar signal in the case of large caps and small caps. Small caps as represented by the Russell 2000 are failing to confirm the higher high in the S&P500. These are events we'll have to monitor going forward to see if they have predictive value.

Hopefully you find some of this useful. My plan is, going forward, to pop in weekly or monthly and give an update of what I'm seeing on the averages. If you're analysis disagrees with mine feel free to comment below with an alternate interpretation. Through discussion hopefully we'll land on the correct answer.