A Summary of Calendar Effects

Eric Falkenstein writes today on the end-of-the-month anomaly. I’ve always been suspicious of these calendar effects.

Some do exist, but they’re often hard to exploit. Or, they have existed but are no longer important.

Perhaps the most dramatic is the days-of-week effect. I looked at the data and found that Monday has been the worst day of the week for stocks by far. Over the last twenty years, however, Monday has done much better. Historically, all of the market’s capital gain has come on Wednedays and Fridays. The rest of the time, the market is net down.

I also looked at the turn-of-the-month effect, but instead of the period Eric discussed, I looked at the last four days of the month followed by the first three days on the following month. That adds up to less than one-third of the total trading days in a year, yet the market has performed at an annualized rate of over 28%. The rest of the time, the market is down.

I recently looked at the First Day of the Month and found that it’s beaten the market since 1996.

Although it’s not a calendar effect, I’ve always been impressed by the market’s correlation to the previous day’s performance. For the time period I looked at (1950 to 2008), the S&P 500’s entire gain had come on days following up moves of 0.64% or more. Half the market’s gain came on day’s following 3.2% up moves which happens less than once a year.

Posted by on April 12th, 2011 at 2:54 pm


The information in this blog post represents my own opinions and does not contain a recommendation for any particular security or investment. I or my affiliates may hold positions or other interests in securities mentioned in the Blog, please see my Disclaimer page for my full disclaimer.