Thursday, September 15, 2016

Dual-Momentum Investing

Part of what drew me to value investing were the statistical studies that showed low P/E (or EV/EBIT, take your pick) stocks have outperformed the market over basically any large time period. This makes it very clear that the inefficiencies that gurus like Buffett talk about are indeed real and in some cases rather easily exploitable. But value is not the only winner in statistical studies. Momentum seems to work too. I'm not about to abandon fundamentals and become some momo guy, but there may be some benefits to learning about what the statistical advantages are with momentum so that in overlapping cases (e.g. deciding when to sell after a value stock has run up), perhaps more profitable decisions can be made. It was with that in mind that I read Gary Antonacci's Dual Momentum Investing.

Antonacci has come up with a strategy that uses both relative and absolute momentum to outperform the market. Unfortunately, it doesn't really apply to individual stocks. Or maybe it does, but if so he hasn't provided any data. He focuses on indexes: buy the index when momentum (as measured mostly by 12-month look-backs, since this is where the statistical studies win) is in you favour, and buy safer stuff when it isn't.

He backtests his methods using index ETFs, but since these are relatively new products (yes, in finance I do consider a several-decade-old product to be new), the tests don't go back far enough to satisfy me. He also criticizes other methods for not having enough of a sample size!

No matter what the tests say, though, the logic behind momo just doesn't resonate with me. You can't get a margin of safety, so you're either wrong or you're right. You're not looking to own a share in a business, you're just piggybacking onto what speculators are viewing as hot, and hoping you can get out before they can. Since I am looking for help deciding on when to sell individual stocks (and not entire markets (!)), I was disappointed by the book. On the other hand, Antonacci does cite a number of studies that do point to statistical outperformance for individual stocks, so I do have some direction to find what I am looking for.

One thing I also didn't like was that it takes the author far too long to state his thesis. I prefer when the author lays out his point for writing from the start, and then spends the rest of the book proving it. In this case, the reader is dryly fed a whole lot of background without knowing how it's relevant to the author's main argument.

If you're interested in learning about momentum and how it can work for you, I think this is a reasonable place to start.


juan said...

Saj, I know nothing about momentum investing, but have been told the book on the subject by Wesley Gray, which will be released next month, is very good. And I think it will deal with stocks, not ETFs.

Veeral said...

Hi Saj, you might be interested in this interview with a momentum investor:

I agree with you about having 'no margin of safety' when you invest using Momentum (or at least that's what I understood after reading about it).