Super Duper Seasonality (Teaser)
One of the unexpected side effects of keeping this blog is that I’ve become a bit of a proponent of stock market seasonality (i.e bullish/bearish biases around certain times of the month, year, etc.)
I‘ve never included a ton in the way of seasonality in my own trading, but by putting others’ ideas to the test week in and week out, I’ve come to understand that some of these plays have been significant and consistent enough historically that they have to, at the very least, be respected.
So (hopefully) before April kicks off, I want to start issuing a (free) monthly calendar of seasonally strong/weak days for the upcoming month that includes: (a) the turn of the month, (b) the day-after options expiration, (c) strong/weak calendar months, (d) the mid-month bulge, (e) individual holiday seasonality, (f) last day of the month weakness, and anything else I can dig out of the MarketSci archives.
I don’t think seasonality is enough by itself to trade on, but just for grins and giggles I’ll be putting $10k into a ProFunds account to follow all the plays and make reporting results easier.
The technical challenge I’m brainstorming at the moment is the best way to report the strength of each of the seasonality plays for any given day. Some observations have been stronger than others and at times they’ve overlapped, so we need a smart way to create a combined map of the month.
This is just a teaser (apologies). I’ll be doing some posts on the new seasonality plays I mentioned above, and hopefully releasing the calendar map before April.
As always, reader input is very much appreciated.
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.
Filed under: Monthly Seasonality Map | 6 Comments
I agree that seasonality by itself it’s usually not strong enough to justify the building of a trading strategy by itself. But it can certainly be a useful tool in combination with other indicators as you do on your SOTM report.
fm
Michael, I too have become more interested in Seasonality as it seems that the effects do persist over time but I wouldn’t be betting the farm on a
“one-off” i.e. that you will always lose in Sep.
I am testing Holidays, EO and Turn of month effects trading NDO.
I have found that applying a Close>200ema filter adds value. Furthermore I found the worst months are Aug and Sep (no filter) and Jul/Aug with the filter.
John
RE to John: hello sir. You can see our monthly seasonality maps here:
http://marketsci.wordpress.com/category/monthly-seasonality-map/
I’m not considering strong/weak months in the map at this time. To your point I don’t think they’ve been consistently strong/weak enough to justify much consideration (relative to other more short-term factors).
michael
Michael, I think you misunderstood me. I wasn’t talking about any particular month being consistently stong or weak in terms of simple returns rather in terms of trading OE week, Turn of Month effect or the holidays. That is to say some months results of TRADING seasonality effects appear to be consistently worse than others.
John
RE to John: aha, got you now. Yes, I like where you’re going – there are some real differences in how some effects have behaved depending on yearly sesaonlity (turn of the month around New Years being a simple example).
My only fear with trying to slice and dice the data too far is in curve-fitting. Generally speaking we’re already dealing w/ pretty small sample sizes when talking about seasonality events. There’s a ying and yang there.
Good stuff.
michael
Michael, I didn’t drill in on the indvidual Seasonal Elements by Month to filter them out rather just summed up all the Seasonal effects for the month an effect became active (not necessarily when the trade thus invoked was closed) and then dropped any months which had -ve returns. Curve-fitting is my bogeyman too no question about that.
John