How to setup an easy momentum system?

Question

Beon · Accepted Answer

Step 1: Select a look back periodWhen we’re looking at momentum, we want to pick a time window, to see what the strength of a stock is over a certain window of time.That can be 3 months, it can be 12 months. Most academic studies tend to focus on 12 months, which is exactly what I do. So that’s what we’re going to focus on today.So select the lookback period. That’s your first step to building this kind of a model. So our look back period is going to be 12 months now.The $64 trillion question is how do we measure momentum? Basically, momentum is the price gain over that period of time and our case over the last 12 months.So that’s called the rate of change, and that’s what I use. But other people use something different. Andreas Clenow, who some of you may have heard, he’s just recently been speaking up in Asia, he uses linear regression as an example.There’s plenty of ways to measure price changes over a period of time. But personally, I use the rate of change.Step 2: Calculate momentum and rankNow then what we do is we measure all the stocks in our universe and then we rank them. And the simple way to do that is via the rate of change for that price periods.So the stock that has the highest rate of change over the last 12 months gets ranked number one. Our research also suggests that you can divide that momentum with the volatility of the stock as well. And again, you would rank it so stocks with the highest rate of change and the lowest volatility would be ranked the highest.Step 3: Equalise volatility (or not)So the next step is to decide how many shares to buy and there are a few ways to do that.One way is to take into account the volatility of the underlying security. And basically what this comes down to is the more volatile the stock, the less you buy and the lower the volatility of the stock, the more of it you would buy.So we equalize the position based on the volatility. Now that’s one way to do it. One of my models does it that particular way.The other way is you can use an equal weighting method. So for example, you might want to have 10 positions and each position is allocated 10% of the available capital. So my other model operates on that basis.Two different models, doing slightly two different things, give slightly two different results.There’ve been various studies that show that the equal weight allocations do actually outperform. This is explained by skew. I don’t want to get too technical on you, but that’s when you get very large positive returns on just a few stocks, which will offset the modest or negative returns of the average stock.So there are these two ways to do it. I do both on the different models simply for diversification purposes. That’s the whole idea. I don’t try and optimize it.Step 4: Align symbol and index trendsNow we’ve all heard of the phrase, a rising tide lifts all boats and a falling tide tends to drop all boats. This is an incredibly important step in the whole equation.Don’t get me wrong. During the GFC, some stocks did rise, but by and large, the extreme majority of stocks fell.This particular rule is to get involved in the market. We want to use a two-step approach. We want to ask the question is the broader trend of the market up? If it’s yes, then we want to invest.If it’s no. We don’t want to invest.And that’s the exact reason why I didn’t participate in a lot of downside in 2008. Some of my systems, as I said before were 100% cash.It wasn’t because I was smart or because I predicted what was going to happen. To put it simply, and this sounds incredibly simple, but it’s true – the market trend was down.I was sitting in cash waiting for that trend to turn up. So if the broader trend is up, I’m involved like I am now. I’m 100% invested for one reason. The broader market trend is up.I don’t know how long that’s going to last. It could last a day, a week, a month, another 10 years. I don’t know.But whilst is up, I’m going to stay invested. If it does turn down, I’m going to get out.There’s a lot of empirical evidence supporting this fact, and again, I stress it and show you a lot of tests in my book Unholy grails.Then the next step, what we want to do here is we want to ask the same question of the individual stock.So our first question is the broader market up or down. Then we say the same thing to the individual stock. Is the trend of this individual stock up or down?If it’s up, yes, we want to invest and if it’s no, we don’t want to invest, and this is where my momentum models get a little bit different from what you can read about in academic papers.Academic papers will continually invest regardless of whether the stock market itself is going up or down, or whether the stock price itself on the individual stock is going up or down. I don’t do that. I want them all to be going up.But a lot of the white papers, they’ll invest all the time. Now, why would you go long stocks that are in a sustained bear market? It makes no sense to me. So that’s that very easy rule.Step 5: Buy new, exit oldAnd the last thing to do is to actually execute these, these rules.Now you can do it weekly. I prefer to do it monthly. I can’t see any evidence that doing it earlier than a month makes any difference. There’s no empirical evidence that I can come up with that shows as a benefit to doing it.So literally what that allows me to do is rank the top stocks after the close of business on the last day of the month.For example, after the close of business on the 31st of December. I’ll run these models. They will measure the rate of change in these markets for the last 12 months. They will then rank those stocks in order of that strength.Then the system will ask the question, is the market going up? And it is, and then we’ll ask the question, are these individual stocks going up?And if they are going up, then I will buy the strongest ones. Simple as that. And then I just repeat that. Every single month. It’s as simple as that. Repeat from step two.So to give you an answer, here are some of the top Australian stocks:Top Australian stocksIt works exactly the same for every other market in the world. Doesn’t matter if you’re in India. It doesn’t matter if you’re in Belgium or the US or Canada to go and be exactly the same.So here’s a long list of the top 20 stocks in the ASX 100 measured by the momentum. And what we’ve got here is using a ranking that divides the momentum by volatility.And in this particular case, for example, let’s say we wanted to have a concentrated portfolio of just 10 positions, and what we do here is we take that top one, which is ticket FMG.It’s ranked number one and it’s saying there that I want to buy 10.54%. Of my portfolio money to buy that stock, and then with AST, it’s obviously not as volatile. I would buy 13.4% of that, and I go down the list until I’ve spent 100% of my money.So the major differences here are:The more volatile the stock, the less you will buy itThe less volatile the stock, the more you will buy it.And that will change on a regular basis and you can adjust that. Personally, I don’t, but some people do. I just buy the position and let it sit until I get an exit.Now, some of you probably just have a question on that. You’re probably going to say to me, well, how do you exit a position?That’s a good question. Well, you exit the position when it falls out of that top 10 rankings.So say for example, here, I buy stocks one through 10 on this list. So if FMG all the way down to QUB, I hold them for a month and the next month I do the same exercise.Now if QUB drops to number 11 and GPT comes up to number 10, I’ll sell QUB and I buy GPT, and I continue to hold all the other ones.You’re just repeating this process and what you’re automatically doing is holding the strongest stocks all the time. You’re never holding a weak stock, you’re always holding the strongest stocks. So we’re always buying that strength.We can feed these rules into a computer. The software that I use is AmiBroker. Earlier on in my career, I use Trade Station, but I moved across to AmiBroker when I started trading equities.I liked the platform, it’s pretty simple to use. It gives you a blank piece of paper, a blank slate, and it does exceptionally good on portfolio testing across many universes of stocks.So what we’ve done here is put the rules into the system back to the year 2000 and we’ve tested it on the top 500 Australian stocks, which is the all Ords accumulation index:Momentum in the top 500 ASX stocksAnd you can see here, if you would follow these rules over that period of time for the last 20 years, you would have had a return of 15.7% versus buying a whole of 8.6%.And you can see in 2008 and 2009 the portfolio flattens and this is exactly what my portfolio does. I just sat in cash for that whole period of time. No pain, no nerves. Just able to sit there.There are 2 important upsides for that. First of all. You don’t lose money. That’s always a good thing. I haven’t met too many people that enjoyed losing 30%, 40% of their money that they had in mutual funds.Yes, some of those mutual funds came good again, but that was a very uncomfortable period of time. Now, don’t get me wrong. I lost money in 2008 and some of my portfolios, one was 13%, others I lost nothing.That put me in a very strong financial position, which in turn put me in a very strong psychological position to pull the trigger again in 2009 when the market started going up.Now a lot of people that lost a lot of money back in the GFC, even today, 10 years later, they cannot get involved. They’re too scared to get involved because they’ve lost so much money.If you have a mechanism to protect the downside, you’re going to be more able to get involved and make money when the market’s running up.So you can see here on the Australian market, it looks pretty good in terms of return. Don’t get me wrong, we’re not hitting the baseball out of the park. That’s what I’m not about.If you were coming here expecting to make 200% per year returns or 15%, 20% monthly returns, that’s not what I’m about.When you’re trading 6 ,7 figures, you just want to chip away with good returns like this over the longer term, and compounding will do its thing.So the other thing we get from putting this data or generating this data is not only we can see the equity growth, we can see the drawdowns, we can see exposure, we can even go down to sector exposure, all that kind of stuff. I can give you all sorts of statistics.For example, I can tell you. That this strategy is only ever had three losing years since 2000 it makes money in 75.2% of all months.The average losing month is 2.9%, the average winning month is 3.8% and the maximum amount of losing months in a row is 4. So with that information, you can arm yourself and have confidence that.You know, one of the things I see is people start my strategies and they have two losing months and they say, Nick, this is not going very well. Maybe I should go and do something else.If you know there’s a chance of having four losing months in a row, well you’re in a better position to just hang out and just let it do its thing.But when you’re trading discretionarily you can’t actually do that because you don’t know. There is no solid evidence of what can and cannot happen. The most important thing, believe it or not, is the drawdown.So with a drawdown, that’s how much your account will decline at any given point of time. Every single strategy on the planet has some kind of a drawdown at some stage.

How to setup an easy momentum system?

1 Answer

1 Answer

How to setup an easy momentum system?

1 Answer

1 Answer

Welcome back

Create Account

Reset password