Large vs Small Cap Seasonality Strategy

05Dec10

This is a follow up to my post re: Large vs Small Cap Performance by Month.

In my first post I showed that large and small caps have historically outperformed during certain months (independent of the impact of volatility), particularly small caps in January and large caps in October.

After tinkering with a few approaches to turn that observation into a working strategy I’ve concluded that only January and October have been consistent enough to potentially justify a trade.

To illustrate, the graph below assumes an investor went long/short small/large caps every January, and long/short large/small caps every October, from 1930. The other ten months per year are spent in cash (with zero return on cash).


[logarithmically-scaled, monthly-interval, growth of $1]

This is a pairs trade where we’re capturing the difference in performance between the two assets, but as discussed in my last post, the two positions are not split 50/50%. Allocation to each is based on trailing volatility (more volatility = smaller position, and vice-versa).

I’ve assumed we traded at 2:1 margin (because of the low-vol nature of the strategy), but ignored margin costs, transaction costs, and slippage.

This strategy wouldn’t be appropriate for leveraged mutual funds (one of our weapons of choice), so those assumptions are obviously not realistic, but even if they were, the strategy turned in only a 3.1% annualized return.

67% of trades were winners and the approach has remained pretty consistent over 80+ years, but there’s just not enough meat in the returns to overcome trading frictions and still justify tying up so much capital.

So is it a bust?

Not completely.

The one place I might still apply the observation is with our monthly Tactical Asset Allocation model.

If the model selects the S&P 500 (SPY, large caps) as a position for January I might consider replacing the position with the Russell 2000 (IWM, small caps).

Given the high correlation between large and small caps, we would be honoring the reason why the strategy selected the S&P 500, while gaining some of the seasonal edge that small caps have exhibited in January.

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.



3 Responses to “Large vs Small Cap Seasonality Strategy”

  1. 1 Mark

    Hi,

    I’m not sure but could the better overall performance of large caps in October not be explained by a few large outliers (i.e. October’s famous crashes)?

    Large caps on average decline less than small caps during crashes…

    Thanks,

    -Mark-

    • 2 MarketSci

      Hello Mark – good question, but no for a couple of reasons.

      We’ve adjusted for the difference in volatility between large and small caps. Even if that volatility adjustment was wrong, October large cap outperformance has been far too consistent to be a result of a handful of outliers. michael


  1. 1 Monday links: implied sales growth Abnormal Returns

Leave a Reply

Fill in your details below or click an icon to log in:

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 )

Connecting to %s


Follow

Get every new post delivered to your Inbox.

Join 35 other followers