Seasonality Map for January, 2011


This is a monthly feature at the MarketSci Blog.

Below is a map of potentially strong/weak days for the US stock market in January based on historical seasonality patterns. Read more after the image.

Scorecard: since launching in April, the monthly seasonality map has called the closing direction of the S&P 500 correct 50% of the time with winning predictions 1.5x losing ones (solidly outperforming 55%/0.9x for buy & hold).

Click for a more detailed look at the real-time performance of the map, and scroll down to the bottom of this post for the new large vs small-cap seasonality map.

About the Monthly Seasonality Map

One of the unexpected side effects of keeping this blog is I’ve become a proponent of seasonality (i.e. bullish/bearish biases around certain times of the month, year, etc). This seasonality map forces me to tie my seasonality studies together every month and is even being used in our own proprietary strategies (read how).

The studies included are: (a) the turn of the month, (b) the first and last day of the month, (c) the day-after options expiration, (d) the monthly W, (e) individual holidays, (f) scheduled Fed meetings, and (g) strong/weak calendar months.

To be clear, I do NOT think that seasonality alone is sufficient to justify a trade; however, all of the seasonality plays included in this report have been consistent enough that I do think they should be one of many tools in the trader’s toolbox.

Some observations have been stronger than others, so I’ve rated each from -100% (most bearish) to +100% (most bullish). Very low ratings (+/- 25%) indicate the play has been inconsistent and should be viewed with an extra skeptical eye.

This is a constantly evolving work and reader input is always appreciated.

. . . . .

New Large vs Small-Cap Seasonality Map

Starting this month, I’m also going to be issuing a monthly large vs small-cap seasonality map (click to zoom). Positive (negative) values indicate large cap (small cap) relative strength.

Note that these seasonality biases only exist after adjusting for differences in volatility between large caps and small caps (aka “market neutral” or “beta neutral”) so they are only relevant to either (a) pairs trading (like our new PWB strategy), or (b) choosing between large and small caps after an investor already had a view on the market as a whole (read more).

The studies included on this map are: calendar month, intra-month, intra-week and individual holiday seasonality.

Happy Trading,

. . . . .

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

3 Responses to “Seasonality Map for January, 2011”

  1. 1 Tom Harney


    Thank you.

    A consideration for the monthly map: adding liquidity of the Federal Reserve’s POMO operations. The POMO schedule is .

    I have been following the Federal Reserve’s operations and the ratio of Accepted to Submitted has been projected as “important” and the average seems to be 4. Below average has projected buying support to the market and above ratios have been projected as no “additional” * effect. Today’s ratio, 12/29/10, is 4.05.

    Total Par Amt Accepted (mlns) : $5,387
    Total Par Amt Submitted (mlns) : $21,802

    * Note – Human beings make up stories thus I am bracketing the words, additional and important to communicate that I do not have data to support these beliefs, just an opinion that asset targeted liquidity helps. Alternatively, I am amplifying a story that Federal Reserve wants amplified.

    • 2 MarketSci

      Hello Tom – thanks for the smart comment. I’ve looked quite a bit at POMO and I (contrary to some of my counterparts in the blogosphere) do not see it as an actionable seasonality event.

      The problem in my mind is that the two major POMO “bursts” have come during very specific market regimes (that happened to be bullish). While anecdotally this seems to support POMO as a strong seasonality event (and it “makes sense”), empirically it’s tough to prove.

      I’ve seen other analysis that look at the POMO days as individual events (i.e. each counting as one observation in the sample). I would argue that because (a) there isn’t a strong “day of” POMO effect, and (b) POMO days have been clustered during very specific market regimes, that in fact the number of observations is only TWO (and two observations do not a conclusion make).

      Just my $0.02.


  1. 1 Wednesday links: shellshocked CEOs Abnormal Returns

Leave a Reply

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

You are commenting using your 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