Fool.com: Diversifying Your Stock Screens [Workshop] May 30, 2000

Workshop Portfolio Diversifying Your Stock Screens

By Moe Chernick
May 30, 2000

Last week I discussed the three major types of stock screens you can find in the Workshop -- Momentum Screens, Value Screens, and Growth Screens. Among the various screens, I also mentioned how some screens select only large capitalization stocks while other screens are not restricted by size. Each of these screens can be used to select stocks for various strategies.

Let's first look at some of the non-analytical reasons for investing in multiple screens:

  1. While all of these strategies are backtested for at least 14 years, it is possible some won't work in the future. Using several different strategies increases the odds that your portfolio will continue to do well even if a particular strategy fails.

  2. For a diverse portfolio, most people would recommend holding 12-18 stocks. Most workshop strategies select the top four or five stocks from a screen, therefore three to five screens are needed to get such diversity.

  3. By using screens that select different types of stocks, when one style of investing goes out of fashion (like when investors prefer large cap instead of small cap stocks or growth instead of value stocks) you will be ready for the new style.
Let's look at the numbers and how developing a portfolio of screens affects them. For this example, I picked the following four screens:

  • Keystone - 5 stocks, annual hold (large-cap momentum)
  • Spark - 5 stocks, annual hold (large-cap growth)
  • PEG-26 - 4 stocks, quarterly hold (growth/momentum/value)
  • RS-26 - 5 stocks, monthly hold (momentum)

I picked these because they would give an investor a nice diversity of both holding periods and stock types. The numbers below are annual returns for each strategy for the period 1986-1998. The Blend assumes an equal weighting of 25% in each strategy. (Note: 1999 numbers aren't included because 1999 fiscal year has just ended [i.e., June 2000] and those numbers are not yet available):
Year     Key  Spark   PEG     RS   Blend
1986     25     26     47     28     32
1987    -15    -13    -23    -33    -21
1988     42     51    117     69     70
1989     23     18     23     39     26
1990     11      8     19     61     23
1991     24     41     12     26     26
1992      5     19     51     66     37
1993     13     13     14    -14      7
1994     24     21     47     52     36
1995     30     32    134     78     69
1996     40     39     44     31     39
1997     47     80     -4     36     40
1998    110      3     75    152     85
CAGR    26%    24%    37%    39%    33%

From this chart, one can draw a number of interesting conclusions.
  1. Yes, you would do better if you had picked the highest returning screen (RS-26, monthly) and just invested in it. If we were assured of history repeating itself, that is what the long-term investor would do. However, history teaches us lessons but it never repeats itself exactly.

  2. Notice how the blend smoothes out the returns between the various screens. However, in the one bad year (1987), the blend could not save you.

  3. Notice the inverse relationship between the returns of the Spark and the PEG for 1997 and 1998 and how the blend gave you two great years despite the bad returns of one screen in each of those years.
I know what you are thinking: This is nice, but times have never been so volatile as the last 12 months. How would such a blend work for that period? I will answer that question next week when I report on a real money portfolio that was started 18 months ago and whose trades and results have been followed on the Foolish Workshop and the Mechanical Investing discussion boards since inception.

Until next time, Fool On!