Change text size

• Email This Article
• Print This Article
• Digg This
• Delicious
• Subscribe

Typically the monthly flow for my articles is mapped out in a way I feel I can wrap things up. This month I may have done that for myself to some extent, but likely leave the reader with a bit of work. Hopefully the articles help distinguish the characteristics of secular bulls and secular bears, without putting too much fear of investing in the latter.

Given my bearish bent and concerns for the year ahead, I had to identify an objective tool to help in my decision-making. That's relatively easy to do since I genuinely believe that anything can happen in the markets and no one knows what the next day or week may bring, regardless of what a forecast may suggest. In April 2009 I never could have imagined probabilities lining up for a close on the S&P 500® Index (SPX) above 1,100.

Testing Long-Term Returns

Over the years I have consistently had good results when filtering systems using the relative position of the 50-day and 200-day simple moving average [SMA], so it made sense to look at that one more time. In order to get the most comprehensive results, I went back to 1915 on the Dow Jones Industrial AverageSM (INDU) using daily closes from a Worden Brothers TeleChart export. I used daily closes so that returns would reflect day to day changes rather than end of month results.

I compared results using No Timing-invested in the market the entire time-and Timing using the 50-day and 200-day SMAs. I also wanted to see if employing a timing tool was more beneficial in a secular bear than a secular bull. Once again I used Ed Easterling's book, Unexpected Returns: Understanding Secular Stock Market Cycles, to identify bearish versus bullish periods (see January 14 & 21 Market Trends articles for more information on secular bull and secular bear markets).

Timing Approach

The Timing approach is invested the following month when the INDU 50-day SMA is above the 200-day SMA at the end of the month. It is not invested when the INDU 50-day SMA is below the 200-day SMA at the end of the month. By using a monthly review rather than a cross of the 200-day SMA by the 50-day SMA, noise from sideways market action (trades in/out of the market) is reduced. It also is more practical for those investors who are not sitting in front of a computer watching the markets daily.

Table 1 provides the results (without dividends) of a No Timing approach versus Timing approach for INDU from 1/3/1916 through 12/31/2009. The data from 1915 was used to generate a 200-day SMA. It also provides some basic trade data which excludes commissions, slippage and taxes.

Table 1: Timing versus No Timing (1916-2009)

Timing results will be negatively impacted by the addition of trading costs including taxes and positively impacted by the addition of risk-free returns via T-Bills when not invested in INDU. In 94 years, 90% of the returns were generated when invested approximately 65% of the time.

While you can't simply subtract trading costs because they should also be compounded, it appears both results could be considered acceptable for an investor. So when someone tells you that an investor who entered the market in the mid-twenties before the crash was able to realize gains by investing for the long-term, it appears the data bears this out. That means if you have an 84 year investing horizon that includes a depression near the start and an 18-year bull market towards the end, don't worry about basic timing.

Table 2 breaks out the periods using secular bull years and secular bear years identified by Easterling. It also assumes that the most recent period from 2000-2009 was a secular bear.

 

Table 2: Balances & Total Returns by Secular Periods (1916-2009) 

While the No Timing approach during bullish periods significantly outperforms the Timing approach, it's the setbacks during bearish periods that prevent me from completely disregarding a Timing approach. Clearly you have to make that decision yourself. It's important to note that not every bearish period yields similar results. The next consideration is to employ a Timing approach during secular bears and No Timing during secular bulls. The only problem with that is it requires some hindsight.

To address the hindsight issue, Table 3 separates out first year returns for each secular bull and secular bear to get a rough feel for how a one year lag in market identification may impact results. It also partially addresses the start date for the market transition since the true beginning for a secular bull or secular bear will not necessarily be January 1st of the specific year.

 

Table 3: Balances & Total Returns for 1st Year of Secular Period (1916-2009)

Once again there is variation in the results for secular bears-you can't suggest using the Timing approach in a secular bear will always be better. Hopefully it provides you with enough information to determine if a Timing approach is something you wish to consider. As a last comment, another element not measured in the test is the number of sleepless nights or worst case entries/exits into the market.

Clare White, CMT
Contributing Writer and Options Strategist
Optionetics.com ~ Your Options Education Site

Questions for Clare? Please visit the discussion board on the homepage of Optionetics.com.

• Email This Article
• Print This Article
• Digg This
• Delicious
• Subscribe

Recent Articles by Clare White, CMT, Optionetics.com

• March 18, 2010
  Market Trends: Boomer Time Horizons, Part 2
• March 11, 2010
  Market Trends: Objective Investing Decisions, Part 1
• February 25, 2010
  Market Trends: Gold Outlook
• February 18, 2010
  Market Trends: Intermarket Analysis, Part III
• February 11, 2010
  Market Trends: Intermarket Analysis, Part II