“You have to be an optimist in this business to survive, you have to bring an original point of view, why you want to do it in a certain way and you need to be able to explain to people in a way that makes sense to them why you are doing what you are doing”
On today’s show I am talking to Tushar Chande, Co-Founder and Head of Research at Rho Asset Management. Tushar is also the author of a number of books on the topic of how to design rule based trading systems as well as having been actively trading these systems for more than 20 years.
Thank you for visiting, now let’s get to the interview with Tushar Chande.
In This Episode, You’ll Learn:
- Why the random nature of the markets attracted Tushar to develop systems for trading these markets
- Background information on Rho Asset Management and it’s philosophy
- How Tushar utilizes behaviors of discretionary traders to make estimates of conviction about their individual trades
- How Rho has designed it’s Altius Program to offer meaningful protection when equity markets go down
“We try to automate all the immediate trading tasks and processes so that we can have tremendous control, consistency and disciplined action at all times.”
- What to be aware of when exploring an Asset Manager’s performance over a 20 year period
- Offense/Defense Ratio, What it means and How to use it to Measure the quality of systems design
“We’ve mostly gone with break out style systems as opposed to a moving average cross over type systems.”
- How Rho’s models have been designed in order to achieve the overall objective of the Program
- What Trend Following Indicators are used for and why
- Exploring the volatility of the equity curve
- The input data needed to run the the Altius Program, the time it takes and when they run the model
“The core advantage of having a CTA in your portfolio is to be able to offset significant declines in the equity and bond markets.”
Resources & Links Mentioned in this Episode:
Beyond Technical Analysis by Dr. Tushar Chande
Learn about the Rho Trend Barometer: Hedge Fund Journal – Rho Trend Barometer Explained (PDF download)
“And all the rules are automated so in terms of day to day we don’t really have to think because that’s all taken care of”
Sponsored by Swiss Financial Services and Saxo Bank:
Connect with Rho Asset Management:
Visit the Website: Rho Asset Management
E-Mail Rho Asset Management: email@example.com
Niels: You're listening to Top Traders Unplugged, Episode number 005 with Tushar Chande co-founder and head of research at Rho Asset Management. This episode is sponsored by Swiss Financial Services.
Introduction: Imagine spending an hour with the world's greatest traders. Imagine learning from their experiences, their successes and their failures. Imagine no more. Welcome to Top Traders Unplugged, the place where you can learn from the best hedge fund managers in the world so you can take your manager due diligence or investment career to the next level. Here's your host veteran hedge fund manager, Niels Kaastrup-Larsen.
Niels: Welcome to another episode of Top Traders Unplugged, thanks so much for tuning in today I really do appreciate it. On today's show I'm talking to Tushar Chande, co-founder and head of research at Rho Asset Management. Tushar is the author of a number of books on the topic of how to design rule based trading systems as well as having been actively trading these systems for more than 20 years. So he brings real unique insight to what it takes to design and run a successful systematic trading program. For those of you who are new to the show I just want to let you know that you can find all of the show notes, including a full transcript of today's episode on the TopTradersUnplugged.com website. Now let's get started with part one of my conversation I hope you will enjoy it. Tushar, good morning it's Niels.
Tushar: Morning Niels, how are you?
Niels: I'm doing very well, thanks. Tushar, before we jump into all of the specific questions that we want to cover today and sort of the structure of our conversation, I thought I would just try and go back a little bit because I think it's important that people, when they're looking at an investment advisor also get to know the people behind the company and some of the thoughts that have gone into where they are. So, perhaps with that in mind if you could just spend a little bit of time just to take us back to your own background and how you got first into the whole world of trading and particular maybe of interest it would be why you ended up choosing the systematic route which is of course, still the smallest part of the investment universe.
Tushar: Well, my background is in engineering. I came from a high power R&D background because I have a PhD in engineering and I used to do a lot of mathematically modeling and a lot of simulations, trying to understand and describe the underlying process. So for me it was a fairly natural way or natural transition to primarily use quantitative methods and therefore the systematic approach was in keeping with my training and my ability to solve problems. Now, what we have to recognize is that when I worked as a scientist or an engineer I worked in the world of cause and effect. That is we believe that there some underlying rules of the road or natural process or phenomena that we needed to understand in order to predict what was going to happen.
Now, what I found interesting about trading or the markets was that fundamentally they were a random situation or a random event or a bunch of random events. That is, there was no fundamental cause and effect. Another way to think about it, for the same set of inputs for a model of the equity markets you can get a very large range of outputs so it's very difficult to just make a relatively straightforward study of the markets and come up with some explanatory variables that are going to work that efficiently or that well for ever and ever into the future. So, that sort of explains my background and also my general interest in trading in particular. And then I'm also quite interested by the fact that there was an ability for you to be original by doing something interesting in terms of your research in order to deal with the inherent randomness in the markets. So, even though the markets were random you could still try and find some rules that allowed you to react to the market differently than someone else. So, sort of a combination of where my training and background came from and the curiosity of somebody that was trying to move or deal with a structured situation versus a predominantly random situation.
Niels: And of course you've also done other things, Tushar. I mean developing systems have been a very big part of your life but I guess also it's worth mentioning that you also spend your fair share of time writing about it and trying to explain to other people how a good system should be designed and how you can actually go about testing these things. Would that be a fair assessment?
Tushar: Yes. I have published many articles over time and I have written a couple of books, one of which has gone into a second edition called Beyond Technical Analysis and has been translated into many languages across the world. So, I think it has been an interesting journey in terms of being able to communicate my understand and models for the markets to a wider audience.
Niels: Excellent. Well, let's not keep away from all the good stuff we're going to talk about today and of course the program we're going to be debating is the Rho Altius program and perhaps you could just give me a brief overview of the program that you run and also just when it started and how much assets under management you run in the program today.
Tushar: Right. the program started in November 2007 and we're running somewhere in the neighborhood of $40M +/- on the program. The program is a medium term trend following program but what's interesting about the program is that we tried to think like discretionary traders and tried to capture many different types of behavior into a single program. So we have different groups of systems that represent different types of (inaudible) with the market. So for example if you were a bank and you said, "Okay we want to start a trading desk." then you'd go out and hire traders from other banks that might have individual niches or preferences or skills. So you may hire somebody that's a good bond trader, you may go out and buy somebody who's a good energies trader, you may go out and hire somebody or find somebody who's good at currencies and so on. And each of these people would bring a different approach to their individual area of expertise and we've already discussed that the markets are random so what we did was, we wanted to follow the approach with something a little different. We said, we're going to use the same rules on all markets so that makes it very robust. So, in a way we're not using different rules for different markets as each of your specialized traders might, but we are trying to capture, even in these uniform rules some of the behavioral styles of individual discretionary traders.
So, for example one of the advantages that discretionary traders have is that they can vary the size of their position that is the size of the initial risk, based on their conviction about the trade. So there's going to be something in the information environment that allows them to change size. So, for example when the new Prime Minister was elected in Japan he was elected on a mandate to pump up the economy and increase the growth rate and the inflation rate. So, this is an external piece of information that a discretionary trader could have used to increase his or her size in say the Japanese Yen or the Nikkei equity market. Now as a mechanical system that uses all the same rules in all the markets, we don't have the luxury of factoring in explicitly a different piece of information for every new trade. However, we have some rules built in that allow us to change the size of the position going into a market. So that's an example of where we've tried to think about how a discretionary trader might approach a problem but we have converted that into some rule based algorithms that we can apply over and over again without necessarily being able to incorporate a specific piece of news from the real world into the models.
Niels: Sure. Well let's just go through some of these more tick box type questions that I think is important for people to get familiar with and that is a little bit about the organization and how you've structured Rho Asset Management. Perhaps you could just go into a little bit of detail about how it's done, who does what and also whether you use outsourced parties to help you in certain areas or whether you do everything in-house.
Tushar: Right. We're based in Zurich, Switzerland and we have from day one used a very high level of automation for our day to day trading, reconciliation, back office and record keeping purposes. So, for example if you use the U.S. regulatory standard as a reference they have various guidelines for how different bits of the trading process and the back office process and the record keeping process should work. So, we've automated all of the mechanical or repetitive aspects of the business as much as possible. But of course there is also value to outsourcing some functions because we get a third party assessment of numbers or analysis of our performance. So, for example our administration is partly outsourced, accounting is partly outsourced, our legal and compliance is partly outsourced but we have a couple of lawyers on our board and so on. So, it's always a matter of striking a balance between doing everything in-house versus doing everything outsourced and we feel that since we are primarily traders and we have a strong regulatory responsibility to our regulators and our clients we're trying to automate all the immediate trading related tasks and processes so that we can have tremendous control and consistency and disciplined execution at all times.
And then of course, even with all that, it's good to go outside for some other things like accounting and legal because they can do it more efficiently there's a distance, a standoff distance, an arm’s length measurement of the performance and of course we're not strictly lawyers, we're primarily traders, it's not our interest, so to the extent that the organization supports trading, we've kept it in-house to the extent that there's areas of expertise and advantages to having an arm’s length transaction we've gone out of house.
Niels: So, in other words I guess you could say that the key functions of research, trading, software development and of course client relations is kept in-house while as you mentioned some of the other functions are outsourced in-full or partly outsourced?
Niels: And when it comes to the underlying strategy that we're going to be discussing today in particular, would you say there is an optimal size of the strategy that you see on the horizon and at that point you may say we need to change the strategy or we may need to do different things differently in order to increase the capacity?
Tushar: I would say that the capacity is well in excess of $1Billion and certainly when we get to that point we may have to add more systems or look at the execution issues in order to get the appropriate control on slippage that we need to maintain the profitability of the system.
Niels: Great. Tushar, you mentioned that the program started in November of 2007 and obviously since that time, it's been an interesting journey not least for people in the C.T. industry. Would you say that the program itself has performed in line with your expectations and obviously, I guess, we need to take into account that many programs have performed quite differently before 2009 and after 2009 and is that something that the Altius program has experienced as well?
Tushar: The short answer is yes, the fund has performed differently after 2009 than before 2009. Has it generally performed as designed? The short answer is yes and the long answer is we wish we had done better. Essentially what you have to think about is the macro environment, the external environment in which the fund had had to trade. So, to give you an idea if you are a fan of Formula 1, you know that the cars are very highly optimized for each circuit. But the circuit is a static object, we pretty much know where all the turns are going to be, now if it rains on race day the performance of the race cars falls off dramatically, why is that? Because even the best drivers need optimal conditions to perform at the world class level. Another analogy I can give you is golfers. If you look at the greatest golfers today or 50 years ago all of them are optimized to perform when the course conditions are perfect and they get to practice on the course for a week or longer so they pretty much know where every hazard is and where every bunker is and where every tree is. Now, what happens with those world class golfers when the wind picks up? Well their scores pick up as well, so if I told you that somebody scored a par for the third round you really can't evaluate that score without knowing how the rest of the field has done and what the environment was. So, on a very windy day, turning in a par score may actually be a phenomenally strong or successful round.
So the same thing happened to the CTA industry and to our program as a whole. Due to the incredible credit crisis and crack-up of Lehman Brothers we had a massive breakdown in the entire world trading system so that the central banks had to intervene and so forth. So, essentially what happened was the markets were working in a particular way, before say 2007-2008 and everybody's systems, including ours, were designed to perform under those kinds of markets where you had maybe a year of two of very weak performance and then very strong trends so you could recover and continue to make new equity highs.
What happened after 2009 or 2008, depending on where you draw the line, was that virtually all of the trends were limited to the equity markets and the bond markets so the people who had very strong positions or exposure to these markets and very slow moving systems tended to do better as a group than everyone else and we are part of that everyone else because we are a highly diversified trader we don't over weight any particular sector relative to another sector we sort of have more or less well or significant exposure to all of the major groups in the market. So, again to go back to the golf analogy or the F1 analogy the trading environment has been unusually difficult for the past five years, just to give you an idea we use the Rho Trend Barometer which measures the percentage of markets that are trending or have reasonable trends at the end of every month. And the breakeven level is about say 42.5% to 45% so if only 25% of the markets are trending then typically trend follows tend to move one standard deviation in terms of return for the month. So, the more months you have above breakeven the more likely you are to be profitable. So, if you look at the five year period ending in say 2009 and the five years after that, what you find is there are many more months below breakeven after 2009 than there were before 2009. So, that just tells you that since inception for our program, more than two-thirds of the months have been below breakeven. So that just gives you an idea of how difficult the environment has been for our program specifically and for diversified trend followers in general.
So, if was raining during every race, if you will, if you want to use the F1 calendar, or it's been a very windy conditions for all the majors, if you want to use a golf analogy so for us the system has done what it could given the very uniquely difficult trading circumstances but of course since we didn't have large exposure to equities and bonds our performance in absolute terms looks worse when you look at managers that are over weighted in those particular sectors. So, overall I think the system has done what it was designed to do.
Niels: And one thing we know from the CTA industry is of course that many managers have very long and prestigious track records but when you look at them of course we also have to be aware that there usually has been quite a lot of evolution and changes along the road, in order to get to where they are today and therefore just looking at numbers historically, I guess, can give you a little bit of a false sense of comfort. Would you say that there are particular ranges of time, looking at your own track record, that we should be aware and where maybe perhaps major research upgrades have happened, just to put that into perspective and so we know what we are dealing with in that sense?
Tushar: Yes, Niels. When you look at somebody that has say a 20 year track record you can be almost sure that what they're doing today in terms of systems and market rates are quite different from what they were doing 20 years ago and that's partly driven by the manager's own research, it's partly driven by the customers because they want us to do research all the time and "Make things better." And then of course there is a natural reaction to what may be happening to the AUM to liquidity as different markets become available to trade or become too thin to trade and so on and so forth. So, when you look at a manager's performance it's good to look at the environment within which the performance was produced using something like the Rho Trend Barometer or you certainly need to ask the question have there been significant changes or even small changes to the track record or to the systems and the portfolio aides and how that has played out over the time and how that correlates with the track record, because for example with exactly the same system I could produce quite different looking track records by over weighting a particular sector or under weighting a particular sector so yes that's something that the user needs to be aware of and certainly needs to enquire about.
Niels: And in as of the Altius program would you say that there are two or three periods that are different because of research upgrades in that case?
Tushar: Well, I would say roughly two periods we've tried not to make too many changes and in fact we started with three systems and now we have six systems and most of the original systems we have continue to be used today with some minor changes. So, roughly I would say late 2010 early 2011 is a good time for us to differentiate our track record because we made some significant changes as a result of various things that happened in 2009 and 2010. So roughly two periods in our track record and I'm sure I'd be happy to provide you with more specific details if you desire, but in general there has been, going back to the previous question, you do have to be aware of when changes were made and what was done and how that altered returns. So that is not a result of applying the same rules across the entire track record.
Niels: Sure. And since we are talking about the track record maybe I could just ask you a few very simple questions just again for the listeners to get a feel for what's inside the track record. Are you able to give us a hint of roughly in terms of the average fee structure that you have in your track record, today what would you say that is?
Tushar: It's very close to the industry bench mark of two and twenty that is 2% of management fee and 20% incentive fee.
Niels: Okay, and in terms of the commissions being charged to the trading, what would you say they are approximately?
Tushar: I'd say about say $10 approximately.
Niels: And of course there's always a downside to the CTA and to any investment and just highlighting those numbers as well, in terms of the maximum draw down, meaning from a high to a low, how has that been in the live trading since November 2007?
Tushar: Right. So, the rule of thumb that we use is that the worst draw down is typically four times the monthly standard deviation so our monthly standard deviation is something in the order of 5.1% so if you multiply that by four we should get something in the range of 20%. So, our worst draw down has been 22% which is back in February 2010 which is just a little bit higher than this (inaudible) but it's in the basic range of say three to five times standard deviation. So, we've done a good job of controlling our risk in this extremely difficult environment.
Now, what does that mean? If you look at many diversified CTAs we're not seeing our worse draw down expand dramatically in the last year or two meaning we have stayed well within this 22% draw down whereas many of our cohorts in the business have had significant increases in their worse draw down by a factor of 50% or 100% even in the last few years. So, I'm an optimist so I like to say that your best month is ahead of you and your worst month is also ahead of you and essentially in 2013 that came true for many programs but we were able to maintain and stay above our worse draw down even in the last year or two.
Niels: Excellent. And just to finalize these short statistics, looking at your average winning month and your average losing month, how does that compare?
Tushar: One of the things we like to do, is we like to use a statistic that we like to call offense/defense ratio which is a ratio of the average winning month divided by the average losing month. What does that mean? You are trying to get an idea for how quickly does the system or the program respond when there are good opportunities in the market? So you want to put on positions, increase risk, when there are opportunities in the markets and your average winning month will go up. Conversely when things are difficult, when there are no trends you want to shrink the number of positions, so reduce your risk in the market so the average losing month will go down. So, naturally the better you are at responding to opportunity or shrinking during adversity then this ratio will be more than one and will be relatively large compared to some people who are not as sensitive to responding to what's happening in the market.
So, for example for us the average winning month is 4.65% the average losing month is 3.42% so to give us an offense/defense ratio of 1.36 in very, very difficult market conditions and that's significantly better than many of the other programs especially when you compare diversified traders with diversified traders some of them are at one or a little bit less than one and it's not that their systems are not very good it just means that the trading conditions have been very difficult but even in these difficult trading conditions you can see that our systems have been able to respond aggressively when conditions are favorable and shrink or reduce our risk quickly when conditions are not.
Niels: And then finally I guess a number that's also somewhat relevant and that is of course the percentage of winning month. Even though I know that the period has been probably a period where the general CTA index had been also suffering from a lack of winning month but do you happen to know how the Altius program has done?
Tushar: Yes, our percentage of winning months is approximately 47% so it's a little bit lower than we would like but the markets have been tough and that's the way it shook out.
Niels: Absolutely. So, let's talk a little bit about the trading program itself and I know you mentioned the structure of it briefly at the beginning but perhaps you could just, in your own words, talk a little bit about the overall structure, the number of markets and sectors you trade and also give an insight to the instruments you've chosen for the Altius program.
Tushar: Okay. Niels, going back to where we were, we are trend followers and we are medium term trend followers. But we want to be a little bit different than the traditional trend follower. Now, in terms of markets and portfolio we trend 44 markets, we are very diversified and with roughly similar weights in all the major sectors so we trade currencies, bonds, interest rates, equities (inaudible) primary sector that everybody does we also trade a lot of commodities so our commodity weights is markets are more than half of the market are commodity markets so we have a bit of a commodity orientation versus an orientation towards finance or financials compared to some of the larger managers.
Now, having given you a sense of what we trade, sort of a diversified portfolio, the number of markets about 44 covering all the major sectors on the major futures exchanges let's talk a little bit about what are you trying to do and why are you doing what you do. So, we believe that the core benefit of a CTA or the core advantage of having a CTA in your portfolio is to be able to offset significant declines in the equity and bond markets that is, declines that last for one month to six months or more, so longer periods. So, we're not talking of necessarily providing a positive offset by that I mean equities are down, bonds are down, we are up, on a day to day basis necessarily, though we've done that from time to time, but we're talking of, say, one month and longer periods when there are sustained declines in the equity markets, sustained declines in the bond markets and then you want something in your portfolio that's going to offset that with positive returns. And historically that has been the role of CTAs and we've tried to make sure that our systems are designed to deliver on that promise that CTAs were designed to provide. So, we're sort of portfolio insurers. So if you're a large manager or a large investor or a small investor and you have some investments in equities like a long only strategy like an ATF or a mutual fund or you actually directly own stocks, and then you also have a portfolio of bonds or directly owned bonds or a bond fund, then you want something that's going to give you an offset if there are going to be prolonged declines in these markets. So, that's where CTAs come in.
So roughly speaking we have three groups of systems, groups one and two are primarily trend following that is they buy strength and sell weakness. They don't move too rapidly because you have to allow the market some wriggle room but we've done something interesting in terms of how much initial risk we are putting in, how we are designing our entries, how we design our exists to allow us to differentiate ourselves and give a little different performance profile than other trend followers in the business or other momentum traders in the business so those are groups one and two and group three as we've discussed is primarily our counter weight to groups one and two. That is that allows us to reinforce this offset function that CTAs are meant to deliver. So that's sort of a quick overview of why we do what we do. So, we do what we do so that we can diversify our portfolio of stocks and bonds, we've diversified by trading in large markets we have robust systems by using the same moves on all markets but we have a different philosophies that can be grouped together into three groups of systems that allow us to react and we've already shown you that our offense/defense ratio is more than one so that tells you that our systems have proven themselves, proven this ability to take advantage of opportunities and expand positions and then shrink them just as quickly when the trends are not there.
Niels: And of course we're not trying to extract any of the secret sauce but I think it might be useful, you mentioned you had six different models working inside the Altius program but I do think it would be useful to try and maybe talk through, maybe not all of them, but some of the key models. What kind of indicators is involved, does volatility play a role? And just to give a little bit more insight as to how you've designed the individual models in order to achieve this overall goal that you just mentioned.
Tushar: So, let's talk about our group one system which are primarily a trend following, long short systems. So, there are a variety of ways you can do this. You can do it using moving averages, you can do it using break out style systems, you can do it using some sort of fundamental model or a predictor model from analytical from fundamental data, you can do it using the term structure of interest rates or you can do it by looking at the structure of the various contracts and the forward contracts versus the current contracts and so on and so on. So, there are many different ways of looking at the market, but fundamentally you have to decide do you want to be long or do you want to be short? That is, do you think prices are going to go up so you want to be long, or do you think prices are going to go down so you want to be short? And then you have to decide how much to risk, should you risk 1%, 2% whatever, 5% or whatever the magic number is, maybe 0.3% for you? So you set some initial risk and then you'll decide what happens if you're wrong? So say you put on a position and then the market does something else, so then you get out. And then you also have to decide what happens when things go your way? You think you're going to go short and the market obliges and goes short, moves lower very rapidly or very nicely, conversely you think it's going to go higher and the market responds by going much higher. So, you also have to decide when you get out when you have a profitable trade.
So, all these complex decisions they all interact with each other and one of the challenges and one of the temptations of the business is you could say that it's very easy to think in terms of having market specific models, so we talked about you could go to a bank and hire a bond trader. So, the bond trader may have a lot of market specific information and may be good at trading information flow. But, if you don't have information flow then how do you design a market specific system? So, one of the choices we could to make up front was to have a series of market specific systems or have a robust systems. So, we've chosen to use the same rules in all markets which means that we don't have any market specific systems, which means that we don't have a system that only trades bonds and nothing else. Now, there are some important reasons to do it this way but that just gives you a sense for our philosophy in terms of what we are trying to do, and how we are going about doing it.
Niels: And you mentioned the choice between, say, moving averages and other types of trend following indicators. Which one did you choose and is there a reason why you that chose one over the other?
Tushar: We've mostly gone with break out style systems as opposed to a moving average cross over type systems. To some extent it's a matter of individual preference but we had two reasons to do it. The first reason is that when the markets are trading at a narrow trading range, that is they're just sort of going up and down, up and down in a narrow range, in that situation a moving average system gives you a lot of unprofitable signals. So, one way to avoid trading during a consolidation or a narrow a price range is to use break out style systems. So that was one philosophical reason to avoid trading in a narrow trading range.
The other reason was that you can be more creative in terms of defining whether you should get in or not. So one of the philosophical questions you can ask it should you focus a lot of your energies in designing good entries or good exits or so on? We've spent a lot of time trying to design good entries, that is if you look at the typical trading system, a trend following trading system it will only have 30% to 35% profitable trades and we wanted to increase the percentage of winning trades so that is why we went with our break out style strategy that we can combine with a small number of conditions, maybe one or two or three to improve the percentage of winning trades. So, for most of our systems if you look at a long term test, the percentage of winning trades is closer to 45% to 50% in that range rather than say 25% to 35%. So, one of the key design features of our program is even though we use the same rules on all markets we've still been able to structure our rules so that across a very broad set of very divergent or very different markets over a very long period of time we tend to get a higher proportion of winning trades compared to the typical trend following systems that you could easily find in the literature. Niels: And is this choice of using price channels rather than moving averages, is that also a little bit of a consequence as to how you want to manage your risk and your use of stops or not use of stops? Tushar: Yes. If you have a moving average system you typically tend to be always long or always short, so you have a large number of positions. Which means that you have to somehow deal with equity curve that is full of markets that are not maybe going anywhere, so that tends to reduce your offense/defense ratio. When you have a break out style system you're essentially changing the nature of your equity curve by saying that we're trying to avoid taking positions in markets that are not experiencing strong moves, so the challenge of managing the equity curve is different because you don't have the continual equity curve composed of lots and lots of position and lots and lots of markets but you're having a discontinuous equity curve with only a small number of positions that can expand or shrink so when the markets are trending you can have a large number of positions, when the markets are shrinking you have a very small number of positions. So, the volatility of the equity curve is not constant or is not relatively stable as you would have if you had a moving average system because in that case you would have a lot of positions on all the time. So, some different set of challenges but on the other hand it's a way to differentiate ourselves and provide something different like a good offense/defense ratio.
Niels: Sure. And in terms of the inputs in your models when you run them every day, what kind of input do you need in order to run the Altius program?
Tushar: We just need the daily price data of open, high, low and close. So for example because we are an algorithmic or a systematic program, that's all the information we need in terms of the daily open high, low, close data. We don't need to have fundamental data or other sources of data or multiple contracts data in order to make our decisions.
Niels: And how frequently do you then run the model in order to implement all these trades?
Tushar: Right. We are an end of day trader, so that means that we only have to run our model once a day after the trading has closed for the day, as opposed to a say if you're a moving average system you might get a cross over in the middle of the day then you have to decide if you want to take it or not, or if you are a very short term trader you may be making a lot of intraday data updates in order to generate new signals, whereas we are really only updating the data once a day, at the end of day. So, we are an end of day trader and we only have to do it once every trading cycle.
Niels: And so what kind of orders do you have to implement in order to run the strategy? What kind of order types do you use?
Tushar: We primarily use three kinds of order types. That is we have spread orders when we roll our positions, we occasionally use market orders to enter and exit position if we have a new account or if we have some sort of trading error but most of the time we're just using stop orders, which means that the price has to exceed a certain level, like rise above or fall below a certain level, called a stop level or stop rise before the trading gets executed.
Niels: And that means, if I understand you correctly, that you don't scale into any positions, you basically want to be getting a full position on when you a signal is triggered?
Tushar: Correct. Again this is a matter of design, not a matter of preference and there are advantages and disadvantages to every approach but the approach we have selected is to be all in or all out. So we have stop orders and the entire position will be put on at one price point, either put on or taken off at a single price point or as close to a single price point as we can get.
Niels: Sure. So, before putting on a position how do you go about calculating, because I guess that is a much bigger part than people are aware of, the position sizing and the use of leverage is obviously a very important part in getting a successful result over time in the CTA industry. How do you go about that side of things?
Tushar: Right. As we were talking about a couple of minutes ago, we were talking about a moving average type system or a break out style system and the difference in the equity curves so our risk control is embedded in our design process because the number of positions we have on is not constant. Conversely if you had a moving average type system where you always had a position in the market, either long or short, then you may have to adjust the positions over time more frequently. In our case we have looked at the long term simulation of the system and determined an initial risk level, at the market level or system level, and the total standard deviation for the simulated results over a very long term horizon and we've combined the two to come up with what gives us reasonable risk control and draw down risk control over the course of the trading.
And all the rules are automated so in terms of day to day we don't really have to think because that's all taken care of in our, what we call ITP or integrated trading platform, but really the data for the individual risk for any one trade, for any one system on any one market and the overall risk is all coming from our simulations which cover a long time period and have various checks and balances for robustness. But more interestingly regardless of what our testing may have been you can see and look at a real time track record and seen that we have controlled our risk very well on a daily basis as well as on a draw down basis for the entire program.
Ending: Ready to learn more about the world’s Top Traders? Go to TOPTRADERSUNPLUGGED.COM and signup to receive the full transcripts of the first ten episodes of the show, and visit the show notes where you can find useful links to other amazing resources. Thanks for listening and we'll see you on the next episode of Top Traders Unplugged.
Become An Insider
Subscribe for free and be the first to receive new and exclusive interviews with the world's top traders. As an insider we'll also send insightful bonus content direct to your inbox.Free Instant Access »
You might also like:
Date posted: 09 Jun 2014no comments