A Note about Large vs Small Cap Seasonality
One quick note about the large vs small cap seasonality biases I’ve been writing about recently (read more: calendar month, intra-month, and intra-week).
As I mention in each post, it’s important to understand that all of these seasonality biases only exist after accounting for differences in volatility between large and small caps (aka “beta neutral” or “market neutral”).
When large caps go up, small caps will tend to go up more. When large caps go down, small caps will tend to go down more. That’s simply differences in volatility. So if you want to predict when small caps will go up more than large caps in a non-beta neutral trade, then predict when the market will go up as a whole (and vice-versa).
That’s not what I’ve been talking about. I’m talking about relative performance after adjusting for this difference in volatility.
That’s why I mention in every post that these studies are only relevant to either (a) pairs trading (like our PWB strategy) or (b) choosing between large and small caps after an investor already had a view on the market as a whole.
I mention all of this again because I’ve received a few emails “debunking” my results. In all cases, the reader failed to use a beta neutral approach.
I suspect that if I received that many debunkings then quite a few other readers are having the same problem. Adjust for volatility good sirs and all will be good and right in the world!
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: Random Stuff | 2 Comments
Michael, there are a lot of ways to normalize volatility. Can you share what your favorite method is or is it proprietary? Thanks for another great year!!!
John
Hello John – for the purposes of these posts I’m keeping it simple.
Calculate the 21-day SD of the daily % log changes for both LC and SC. Then use simple algebra to calculate what % you’d need to invest in each to make those 21-day SDs equal. This isn’t the best way to go about it necessarily, but keeps things simple for these tests.
michael