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.

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.

20130531.01

Real-time results since I began sharing the calendar in October of last year have run inline with the historical test.

The S&P 500 has averaged 0.10% (30% annualized) on the best half of days versus 0.06% (17% annualized) on the worst half.

Quartile 4 days (the worst of days) have been particularly bad, with an average return of -0.15% (-32% annualized).

As I stressed when I introduced the concept, day-of-month seasonality never justifies a trade all by itself, but I do think it deserves to be one of many tools in the trader’s toolbox.

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.


As noted by Paststat, yesterday marked the Dow Jones Industrial Average’s 18th straight winning Tuesday. How rare an event is that? Very…

The chart below shows the longest active day-of-week winning streaks for the DJIA since 1900 (regardless of the day of the week).

20130515.01

The red arrow marks our current winning streak of 18. It hasn’t been since the 1920’s that we’ve seen anything close when the DJIA posted multiple 15 day-of-week winning streaks.

I should note that had we looked at the S&P 500 (which for the purpose of an analysis like this is more or less the same thing as the DJIA) the streak stands at just 8 Tuesdays, a much less impressive feat.

Of course this is all just polite dinner conversation and chart porn, and I put no stock in this as a useful timing indicator.

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.


There has been an abundance of “Sell in May” analysis out of the quantitative blogosphere this week. A quick roundup:

Turnkey Analyst

Doug Short

Bloodhound Exchange + follow up

Woodshedder

Paststat + follow up

UK Stock Market Almanac

MarketSci + follow up

– Plus 99% of posts from the Stock Trader’s Almanac.

Hat tip to The Whole Street for all links. Feel free to comment with any I might have missed.

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.


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.

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.

20130430.01

Results in April were (beyond) abysmal, but real-time results since I began sharing the calendar in October have been mostly inline with the historical test.

The S&P 500 has averaged 0.09% (24% annualized) on the best half of days versus 0.05% (14% annualized) on the worst half.

Quartile 4 days (the worst of days) have been particularly bad, with an average return of -0.17% (-35% annualized).

As I stressed when I introduced the concept, day-of-month seasonality never justifies a trade all by itself, but I do think it deserves to be one of many tools in the trader’s toolbox.

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.


Inspired by Hyperq (hat tip The Whole Street)…

The graph below shows how each day is weighted in a 10-day simple moving average (grey) versus exponential moving average (red). For the uninitiated, the SMA and EMA are the two types of moving averages most commonly employed by traders.

20130428.01

In the 10-day SMA, each day from 0 (the most recent) to 9 (the most distant) is equally weighted (10%).

In the EMA, day 0 makes up 18.2% of the average. That falls to just 3.0% by day 9. The left tail on the graph (day 20 and beyond) extends indefinitely, but in total makes up just 1.8% of the average.

The next graph show the daily weighting for a 50-day SMA/EMA, and the graph below that compares a 10-day EMA to a 50-day EMA.

20130428.02

20130428.03

For those familiar with how these averages are calculated, none of this is new information, but I thought it was interesting to see it visually.

For me personally it’s a reminder of how arbitrary an SMA is, weighting the very last day in the average equally with the most recent day, and how much of an impact it can have when that last day falls out of the average despite not telling you much about what’s going on today.

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.