Objective research and reviews to aid investing decisions
 Looking for information on debt consolidation loans? Want to know more about reverse mortgages? |
Can investors beat the market by buying when a stock index crosses above a moving average and selling when it crosses below? In other words, do such crossovers and crossunders signal economically important market turning points? Using daily closes for the S&P 500 index for 1/2/90-7/17/07 (and 1/2/62-7/17/07 for one analysis), we find that:
The following table summarizes the results of buying/selling on moving average crossovers/crossunders for the S&P 500 index using various averaging intervals over the period 1/2/90-7/17/07.
The "Signal" is the length of the moving average interval in trading days. The selected intervals correspond roughly to two weeks, one month, one quarter and one year.
"Round Trips" is the number of buy/sell pairs generated by crossover/crossunder signals. We assume the investor/trader looks at no data prior to 1/2/90. If the index is above its moving average after the first averaging interval, the investor buys at the close. If it is under, the investor waits for the first crossover. All four cases are in the market on the 7/17/07 end date.
"% Days In" is the percentage of calendar days each moving average buy/sell rule puts the investor/trader in the market. In general, the shorter (more sensitive) the averaging interval, the less time in the market.
"Total Return" is the cumulative return from the rule assuming all buys and sells are at daily closing levels and that all cases sell at the close on the 7/17/07 end date. The calculation assumes that the return on cash while out of the market offsets trading costs. It ignores taxes. In general, total return grows with the length of the averaging interval.
"Buy-and-Hold Return" is the cumulative return from buying on the same date as each rule first buys and holding until 7/17/07. The slight differences result from different first-buy dates for the four rules.
Buy-and-hold beats all four moving average rules based on raw cumulative returns. However, the difference is not extreme for the 252-day averaging interval. Assuming that there is a continuously available vehicle to hold the index, buy-and-hold also has tax advantages over the trading rules.
The following chart offers a more robust comparison of buy-and-hold and the 252-day moving average trading rule. It shows the growth of an initial $1 investment in each starting with the initial trading rule buy on 1/24/91 and ending 7/17/07. In general, the trading rule underperforms during bull markets and outperforms during the bear market. Specifically, the trading rule avoids most of the steepest declines during 2000-2002. The rule does generate some clusters of almost daily buy/sell transactions that would be frustrating for a long-term investor, but ignoring some signals would risk missing a critical one.
There are only 17.5 independent 252-day intervals in the above sample. Would a longer sample with a different mix of bull and bear markets yield different results for the 252-day trading rule?
The next table shows the results of a similar analysis for a 252-day averaging interval over the much longer period 1/2/62-7/17/07 (45.5 years, longer than most investing lifetimes). In this case, the rule of buying/selling on moving average crossover/crossunders beats buy-and-hold on a raw return basis, but the difference is not extreme. Again, the rule-based return series is likely less volatile than that for buy-and-hold.
Note that for this longer sample period, there was no continuously available vehicle to trade the index.
In summary, moving average crossunders/crossovers may have some practical value as buy/sell signals for long averaging intervals, generating comparable returns with lower volatility. However, the tax consequences of trading could more than offset that value.
See Blog Synthesis: Some Trading Indicators for analyses of the usefulness of other technical indicators.
