Getting Your Portfolio in Better Shape

Industry Models

I have several industry rotation models, both short and long term. My main one is in the back of my head as I analyze where the pain is growing and nearing maximum intensity. (If anyone wants me to share shorter-term models, I can, although I don't use them much, though.)

For me, the point behind industry rotation is to find stocks that fit one of two paradigms: 1) strong companies in troubled industries and 2) well-run, cheap companies in industries where favorable trends are overdiscounted. My main quantitative model attempts to address the former.

This model has six-and-a-half years of Value Line industry-rank data and asks the following questions:

1. What is the industry's current rank?
2. Relative to your rank history, where is your current rank compared with the maximum and minimum ranks?
3. How many standard deviations are you above or below your average rank?
4. What percentile is your Value Line rank compared with its percentile on prior dates in history.

The results from these questions are weighted and turned into a grand rank. The weights are based on the uniqueness of the information contained in each question. From highest to lowest, the weights go 1, 4, 3, 2 for the questions listed above. This spreadsheet lists the final results. The new ranks can be used in two ways, in value mode or momentum mode.

Value Line ranks are a product of three factors: price momentum, earnings momentum and analyst surprise. They are momentum driven. My model attempts to refine that and give investors two ways to play the market: the fast-momentum style, buying companies near the bottom of the list, or, if you're like me, buying the companies in dead industries at the top of the list.