Turn-of-the-Month Strategy
My recent posts about stock market strength around options expiration week left out two important parts of the month – the beginning and the end. In this report, I’ll share a strategy for trading the turn-of-the-month. This observation will be added to the State of the Market report.
I’m following in the footprints of giants. The turn-of-the-month has been discussed on many occasions by others – notably by CXO Advisory here, here, and here.
[logarithmically-scaled]
In a nutshell, the market has generally been bullish around the beginning and end of each trading month. By my count, this turn-of-the-month bullishness usually covers the first three and final four trading days.
The graph above shows the results of a trading strategy long the S&P 500 on these turn-of-the-month days (green) versus both buy and hold (blue) and a strategy only long non-turn-of-the-month days (red) from 1970 to 11/2008.
Note that this is a proof of concept so this test is frictionless (no transaction costs or slippage) and does not account for return on cash. For the record however, these results could be duplicated using actively traded mutual funds such as those from Rydex or ProFunds.
And for the number lovers:
Clearly, these seven turn-of-the-month days, despite including only a third of all days, are far more bullish than the average day. Such a turn-of-the-month strategy would have outperformed the market in terms of absolute and risk-adjusted returns and significantly reduced downside volatility.
More turn-of-the-month statistics
The following chart shows average daily returns for the S&P 500 on the first seven (left half of graph) and last seven (right half) days of the month from 1970. Geek note: these results have been de-trended by subtracting the average daily return of all whole months.
The bold red line represents an average of all observations and the lighter red lines individual decades in the sample. On average, immediately after the first three days of the month and immediately prior to the last four, the market has dipped (leading into the bearish weeks prior and after options expiration).
Last thoughts
As I wrote previously, I’m not a big fan of seasonality plays, but both of these have been strong enough and consistent enough that I will be adding them to the State of the Market report. I wouldn’t trade either alone, but I would consider them as one of many, many things I’m looking at.
In a follow up post, I’ll combine this turn-of-the-market strategy with the options expiration week strategy, effectively canvassing the entire month. More to follow.
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: Time-based, Trading Strategies | 8 Comments
-
Subscribe to RSS Feed
Subscribe via Email
My email: michael @ marketsci.com (remove spaces)
In my continuing efforts to be a pest – it looks like, on the end of the month, that you’d need to buy on the close of day 5 as it is marked on your chart, and then sell the close of day 2 or 3. Am I reading this correctly?
RE to d: my brief style of writing strikes again =| the returns are not cumulative in the graph, so actually days 2 and 3 prior to the end of the month are still adding to gains. Does that make sense?
great work again…..and i agree, combining with options expiration would be valuable. note that the 1st of the month has the best equity curve and this seems to work on all indices—this was referenced by McClellan: http://www.mcoscillator.com/reports/special/SP_500_FirstDay.html.
as a side note, you may want to look into combining this with monthly seasonality ie Nov to May etc.
dv
RE to david: thanks David. My results above (second graph) confirm that the first day of the month has historically been the most bullish of the seven in the turn of the month strategy. RE: monthly seasonality – as I wrote, I’m not a big fan of seasonality and as we move farther and farther up the chain (daily to weekly to monthly, etc.) the less powerful I think it becomes. I think that there are so many very simple ways to trade that buck the “sell in May and go away” indicator that I don’t think it’s even something active traders should consider. Just my $0.02.
Thanks,
michael
Excellent study…it appears to corroborate one of Larry Connors/Kevin Haggerty core strategies, (i.e., end of month bullishness) which they have advocated for several years.
as an alternative to monthly seasonality which can be noisy, i think that breaking down returns by calendar quarter would be very valuable—–the quarterly effect (q1 ,q4 strength etc) seems to occur in every asset class–making it a more robust indicator. There is also more logical explanation for quarterly differentials vs monthly differences in return.
dv
not to be pedantic, but could you please remove the % behind indications of Sharpe ratio and display it using the standard convention. It may help those unfamiliar with the metric.
RE to nick: Sorry, but no because (a) when numbers line up aesthetically it appeals to my inner geek, (b) I am not a slave to convention, and (c) anyone who can’t figure out that 1.5 = 150% probably doesn’t understand anything on this blog anyways, so the point is moot.
[in all seriousness, I go back and forth all the time on this, and I really do prefer the %, but point well taken]
michael