Andreas Clenow, CIO, ACIES Asset Management
"People think that you can just take a trend following model and apply it to stocks, and frankly, you cannot."
Automated Trader: Why did you focus on momentum strategies for this book?
Andreas Clenow: This is actually my second book. My first book was on trend following, and I got known as the trend following guy, which is fine but it's a small part of what my firm is doing.
What I tried to do in the first book is explain what really is going on behind the scenes of managed futures funds, what almost the entire CTA business were doing.
Trend following isn't all that complex. Anyone can find basic rules on the internet. What are these rules really? The time frame for instance, do you want to follow shorter or longer term trends? What asset classes do you focus on? What risk do you take? These are really the main points. Otherwise, trend following is more or less the same. There's not that many ways it can be done.
So, for this book, why stocks? People are often confusing stock and futures markets, and there is a lot of confusion in terms of trend following, which doesn't really work on stocks.
The whole idea of trend following is based on diversification. You have markets behaving differently; gold, oil, Nasdaq, bonds, currencies - so many different things in the world. The grain markets for instance doesn't correlate to oil, or to the Japanese stock index. Trend following works because most of the time you are losing money on most markets, but a few things will make so much money that it accounts for it, and more.
People think that you can just take a trend following model and apply it to stocks, and frankly, you cannot. The underlying mechanics are not there. You do not have the diversification.
What you can do, however, is reasonably close - momentum trading. So this follow up book explains what is momentum trading, why is it different?
It's something I have been doing for about 15 years or so. Momentum strategies work very well over time. I explain all the rules and enough details that anyone who has enough knowledge about the business, about the technical side and access to the data could replicate it. If anyone wants to criticise what I am doing, I am very happy to hear it. But I am expecting them to do homework. Simulate and see what you get.
eBay: Example of a momentum trade. Position was entered after a period of strong performance. The stock was kept while it was moving up, and the position size was changed over time to maintain approximate volatility parity. As the stock starting losing steam in November of 2012, it was sold in favour of a higher momentum stock.
AT: Maybe one thing you didn't elaborate on in the book is the technological aspect. What does setting up look like in this regard?
AC: I come from the hedge fund side, so we are playing with tools and toys that aren't available to most people. But in terms of accessibility, I see two very different streams of development. It will be interesting to see what will really happen here.
On the one hand you get better and cheaper tools on the desktop side. One of my personal favourite tools is C# application called Right Edge. Very powerful, very flexible application. You can customise, you can rebuild it if you are a programmer, or if you want to build your own add-ins, change your application behaviour.
The other development stream I see is complete on-line solutions. A couple of years ago, I was very sceptical about this, I was asking the founders of one of these major sites why they want to put simulation environments into web browsers and make it worse.
But I have to say, they are achieving some interesting things.
AT: Which sites do you mean exactly?
AC: There are two major players, and a whole bunch of others. The biggest and best is Quantopian. And then you have Quant Connect, which is a smaller start up.
What's interesting is not just the simulation environment, but the fact that they hook up the data.
If you want to get access to a simulation environment, it will cost you not just a lot of money, but a lot of time. You can't just press a button and buy a package of things. You buy a simulation environment, your data, a database, build mapping tables - it is complex stuff.
Quantopian and QuantConnect have solved a lot of things that are very difficult for professionals, such as cash dividend adjustments. Anyone can download unadjusted stock data. But for 10 years? It's not going to be correct.
One thing I point out that is not entirely popular is that if you really want to get into systematic trading, you need to understand technology. The old school thinking of 'I am a trader, I don't need to deal with this' is a little bit to me like a doctor or lawyer, who back in the 70s had a secretary so didn't need to type. But can you imagine a lawyer now who doesn't know how to use a computer? She'd be a big helpless baby.
I have quant people working for me, fixing things, but if I didn't understand it myself, I would be helpless. Step one of getting into this business is learning to programme.