Momentum trumps efficient market
Posted on December 8, 2007 by Richard Beddard
Filed Under Investing |
A buy and hold investor touts momentum trading after examing data compiled by Professor Kenneth R French, long-time collaborator of Professor Eugene Fama, architect of the Efficient Market Hypothesis. EMH, in its strictest form, is the theory that states that investors ought not to be able to beat the market using the value effect, the momentum effect, or any other method for that matter.
How’s that! Two conundrums, in two sentences.
Comments
11 Responses to “Momentum trumps efficient market”
Leave a Reply
Hi Richard,
I just thought I would say that now is not the time to start doing momentum investing! Momentum works in both directions and right now it is down or fairly flat.
I would even go so far as to say that now is a great time to look at value investing, absolute return funds or cash.
Hi Robin, good to hear from you. How are you getting on with James Montier’s book?
I’m no momentum investor so I can’t really comment except insofar as you’d need a market with momentum in order to invest in it (which is what you are saying). However although the stockmarket in general doesn’t appear to be in a firmly established trend, markets in individual shares might be…
Hi Richard,
I delve into Montier’s book fairly often. I find it a difficult book to read because it is so disjoint. There is no flow to help you organize the facts in your mind. Some of the individual papers left we thinking ‘so what?’ or lacked details on key information. Having said that, there are also a lot of very interesting ideas in it (such as the result on low 10 year P/E’s).
I think that the problem with the term ‘momentum’ is that it is very vague in meaning. Just looking at one aspect – time - a day trader will measure momentum in minutes, Professor French measures it over 11 months and I measure it over 5 years. What you should actually measure is another topic entirely.
I think that the main thing to remember with any mechanical strategy is to build in diversification because you can be sure that it will not work all the time.
i’m agree
Hi Parcobuoi,
I am not sure which point you agree with (hopefully all
) however if it is the last point then all I can say is that, right now, my portfolio, while quite diversified, is being completely trounced by the current market conditions.
I do not know how this can be avoided as we are facing systematic risk conditions. Because my mechanical strategy looks at longer term trends, it has no specific strategy for dealing with highly volatile market conditions. In this case I basically have to sit through it and see how it comes out.
Hi Robin, a belated happy new year!
Does it need to be avoided? So long as you’re right about the long-term trend
Hi Richard,
A belated happy new year to you too. I hope you are well.
The falls in the market, from my perspective at least, are no worse than May 2006. In fact they have not even reached those levels yet. From a more fundamental perspective, the main developed markets have fallen a lot further than before which cannot be good. Five years of cheap credit may finally be catching up with us (although better now than later). It is very depressing watching a year and a half of hard work disappearing down the drain in just a couple of weeks.
I am reading Beniot Mandlebrots book ‘Misbehaving Markets’. It is quite interesting. Like Nassim Taleb he rubbishes EMT and, like Taleb, does not offer very much in place of it in terms of solutions. However, he does offer building blocks, thoughtful discussion and ideas. He also has a much more pleasant and engaging writing style than Taleb.
He mentioned that stock returns have a ‘memory’ so that the returns on day one will have an impact on the returns on day two and three. This is clearly not EMT and I thought I would test this using about 50 years of S & P 500 data.
Interestingly, the actual distributiuon of one day returns seem to follow a normal distribution over time with a slightly positive average (which explains why the markets go up perhaps). However, if you try to simulate the index behaviour by randomly picking values from the distribution, then the curve looks nothing like a real market. In a real market, the selection of a return on day one clearly influences the selection of the next return on day two (particularly when the markets are volatile). This means that the biggest negative return over several days is much greater than if you pick the returns randomly.
Anyway, I look forward to your future postings.
Hi Robin,
If you believe George Soros, it’s sixty years of cheap credit coming to an end!
I was inspired by Mandelbrot’s book though, unlike you, I didn’t have the wherewithal to test out any of his assertions myself. Sadly he doesn’t share the fractal maths that enables him to simulate markets (not that I’d be able to do anything with it!) but it’s interesting that you have found what he showed - that you cannot simulate a market with a normal distribution.
Incidentally I don’t know if you saw this, but here’s a recent article in which Taleb and Mandelbrot sing from the same hymn sheet: http://www.ft.com/cms/s/2/5372968a-ba82-11da-980d-0000779e2340.html
Perhaps Mandelbrot is the acceptable face of Taleb
That is a great link and I had not seen it before. How are you finding Taleb’s book by the way?
Mandelbrot’s formulae are all at the back of the book. It is slightly disconcerting trying to tie them back to the chapters. I have not tried to generate a fractal stock curve from the information in the book, however if it is anything like generating a fractal mountain then it should not be too difficult
If I had more time I would try to work it out for interest’s sake. However I have read nothing yet which suggest it would be helpful to my investment method. I already know that all my investments can simultaneously blow up in a bad way. It has happened three times already in a year and a half. Staying invested and making gradual momentum based adjustments is the only thing I know how to do!
Your reply made me laugh out loud! I forgot the equations were in the back of the book - probably because they were gobbledegook to me.
Your final point is the one I made in an article I wrote after I’d interviewed Mandelbrot. Investors (i.e. practitioners, not academics) have always known that markets are ‘wild’ not ‘mild’ (in his words).
I guess we’ll have to wait for book 2 to find out how we can use fractals to predict, rather than simulate markets, but he’s getting on…