Day of Month Seasonality for November

30Oct12

A topic recently on me noggin has been day-of-month seasonality (read more, more, and more). Using a simple walk-forward test to minimize hindsight bias, I showed that trading the days of the month that have been strong historically has consistently led to much stronger returns in the future. That’s as true today as it was in 1950.

The “strategy” would have been uber effective in October, with the S&P 500 closing up on 64% of qualifying days, versus just 33% of non-qualifying days.

Below is the DOM seasonality calendar for next month, broken out by quartiles (read why), with quartile 1 indicating the strongest days and quartile 4 the weakest.

I’ve highlighted just one exception: 11/21, the day before Thanksgiving.

Even though that day technically qualifies as quartile 4 (the weakest of days), the day before planned exchange holidays has historically been bullish for equities, so I’m leaving it as a question mark.

Happy Trading,
ms

. . . . .

To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.



5 Responses to “Day of Month Seasonality for November”

  1. 1 Tjackson

    I really appreciate this Michael. I made some money this month from this and hope this is a monthly feature. Thanks again.

    • 2 MarketSci

      Hello Tjackson – glad you found value in it – I’m traveling at the moment, but I will make time to release one for next month in the next day or so. Be on the lookout. michael

  2. 3 Quintuitive

    Hi Michael, enjoying your posts as always! I have been trying to reproduce your results using R, and have encountered a few problems. First, the 20 year average returns for the second day of the month, rarely have it in the top 5 (quartile one). Are you upgrading it for another reason or is that what you get? The second issue is with events like Sandy last month – the market was closed for 2 days. Of course, when preparing the map for October, one can’t tell in advance, but then Oct seems to have more trading days when it participates in the history for the next 20 years. What returns do you use for these days – 0s? Same for holidays? Doesn’t that bias the returns somewhat? Thanks!

    • 4 MarketSci

      Hello Quintuitive – I’ve mentioned in previous posts (but should be repeating each time I post this calendar) that I’m using a slightly different approach than the simple one provided on the blog here:

      http://marketsci.wordpress.com/2012/10/17/excel-workbook-day-of-month-seasonality/

      First, I’m trying to measure the distribution of returns (rather than just a simple 20 year average) to judge best/worst days. Second, I’m doing a bit of smoothing between days, so as not to be too specific with something that is inherently not so specific.

      Results are going to be very similar to the Excel workbook provided, but as you’ve noted above, some days will be a little off.

      michael

    • 5 MarketSci

      P.S. always good to have other bloggers whom I admire around here. Welcome!


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s