Objective research and reviews to aid investing decisions
 Looking for information on debt consolidation loans? Want to know more about reverse mortgages? |
One of the stock market anomalies we seek to exploit in our Strategy Test is the Turn-of-the-Month Effect, a significant concentration of excess stock market returns around the turns of calendar months. Precisely which trading days produce the excess returns for the principal trading vehicle in the test, the iShares Russell 2000 Value Index Fund (IWN)? Using daily adjusted closing prices for IWN for the period August 2000 through March 2007 (79 months), we find that:
The following chart shows the average daily raw returns, with one standard deviation error bars, for trading days around the turn of the month. Average daily returns range from -0.26% for the seventh trading day of the month to +0.34% for the first trading day of the month. The average daily return for all days in the sample is 0.06%. The standard deviations of daily returns range from 0.96% to 1.28%, so all differences between average returns for specific trading days and the average for all trading days are small with respect to variability.
Visual inspection indicates that excess returns concentrate in the last five trading days of the month and either the first two or first four trading days of the month, suggesting excess return windows of either seven or nine trading days. The average raw (excess) return for the interval [Last-4 through 2nd] is 1.53% (1.08%), with standard deviation 2.77%. The average raw (excess) return for the interval [Last-4 through 4th] is 1.54% (0.96%), with standard deviation 3.30%.
In summary, historical data from the past 79 months suggests a turn-of-the-month effect for small-capitalization value stocks concentrated in the last five trading days and the first two or four trading days of the month.
We will keep these results in mind as we execute our Strategy Test.
Note that the sample size is modest and the noise-to-signal ratio is roughly 3:1, so: (1) the existence of the effect is somewhat suspect; and, (2) if it does exist, exploiting it reliably requires many months of trading. The largely out-of-sample analyses described in our blog entries of 7/20/06 and 8/9/06 support the existence of a broad turn-of-the-month effect.
For related research, see Blog Synthesis: Calendar Effects.
