My Introduction to Timing Models

I was first introduced to the concept of market timing models in the mid-1980’s after a co-worker regularly brought me Martin Zweig newsletters to read. Eventually I picked up Zweig’s book Winning on Wall Street in which he introduced his readers to a simple model that included both monetary and tape indicators. I was immediately hooked by his scientific approach and saw models as a logical way to gauge market prospects. Most newsletters of the time were full of hype, speculation and subjectivity.

Eventually I went on to read classics like Norman Fosback’s Stock Market Logic and kept clippings on anything concerning market indicators. Keep in mind this was an era before most folks had home PCs, let alone internet access – Barron’s and newsprint was as good as it got.

I soon understood market indicators fell into one of four categories:
Tape – trend, breadth, and momentum (velocity) indicators
Monetary – interest rates, liquidity, and economic factors that influence liquidity
Sentiment – surveys, asset flows, P/C ratios, volatility measures, etc.
Valuation – P/E ratio, Bond/S&P dividend ratio, etc. (Ned Davis would say valuation measures are basically the same as sentiment)

I also spent many years investigating technical analysis tools and techniques, but came to the conclusion that most practitioners where either fooling themselves or others. This isn’t to say I didn’t see value in using technical indicators, but saw TA primarily as a way to reduce drawdowns, not as a means to achieve outsized profits. The folks I read about that really excelled at “TA” are traders like Linda Bradford Raschke – who have essentially developed a keen awareness of subtle anomalies that can arbitraged. They aren’t chart readers or people who developed complex timing systems.

I was also a big fan of Gerald Appel, the author of the old Systems and Forecasts newsletter and inventor of MACD. After reading one of his books (I believe it was called Stock Market Trading Systems) I thought he was a genius. As the years went by I began to realize the methods outlined in his book were more of an exercise in data-mining and curve fitting than a system that could hold up over a long haul.

Back to indicator categories: Although I believe monetary and valuation indicators provide great value in ascertaining market risk, they provide more value to investors who have an orientation to typical bull/bear cycles of 3-4 years. The most recent occurrence of the inverted yield curve is a great example. How many months has the market produced excellent returns when this proven indicator signaled caution? Most investors lose faith in indicators that take many months before they fulfill on their message.

My orientation is intermediate term (typically 6-8 month cycles). Using this timeframe, tape and sentiment indicators provide the most value. In my next post I’ll try to outline the components of my timing model and the underlying rationale.