Singapore-based Quant Asset Management was co-founded by Frank Holle and Chatchai Ngampakdeepanich.
QAM manages the QAM Global Equities Fund and the QAM Asian Equities Fund. Both funds were launched in April 2004. The QAM Asian Equities Fund has returned 34% since inception with an annualised return of 37% and annualised volatility of 23%. The QAM Global Equities Fund has returned 29% since inception with an annualised return of 32% and annualised volatility of 16%.
QAM currently has US$17 million in assets.
- Quant Asset Management is relatively new; how did it get off the ground?
- Could you tell us more about the
key decision makers in the fund?
Chatchai is 37 years old, has four children
ranging from 7 to 13 years old and has
worked with Thomson Financial for over
eight years. In these eight years Thomson
acquired Datastream, I/B/E/S and First
Call and became truly the world's biggest
financial database. Chatchai knows the
database inside out and spends a lot of
time improving on it.
- Can you explain a little more about your focus on automation?
- And the investment methodology?
- What is the main theme throughout the investment process?
- In what market conditions does the model work best and worst?
- How scalable is your strategy in terms of AUM?
- What are the countries covered in the Asian and Global Fund?
- How do you see yourself different from other fund management companies?
- You mentioned you will be able to manage US$800 million. Are there plans to launch more funds in addition to the existing Asian and Global funds?
- What are your future travel plans
to meet with prospective investors?
So far we haven't been marketing aggressively but I just came back from a trip to Europe. We'd really prefer to let our performance speak for itself and earn our place in the various rankings. In order to get QAM on people's radar screens, we have listed ourselves with hedge fund databases like Eurekahedge's and we speak at some conferences. I was on such a conference in New York in the first week of April.Furthermore we are planning trips to Amsterdam, Zurich, Geneva and London at the end of the summer this year to update investors on our progress. Since investors often have to visit a manager's office as part of their due diligence process, we prefer them to come Singapore and see us for the initial meeting where we actually explain our methodology.
We set up Quant Asset Management in Singapore in November 2003.
I met my current partner, Chatchai Ngampakdeepanich in early 2002 whilst he was working at Thomson Financial. I had approached Thomson because I was interested in their quant data and Chatchai came to see me in KL to show me their product. After a couple of meetings he showed me some work he was doing in his spare time. For many years he had been researching which factors and factor combinations have an impact on share prices. He had also developed a methodology whereby these relevant factors would regularly get different weightings depending on what actually happens in the market.
His university background is in computer engineering and finance, which is a pretty useful combination and soon it was clear to me that I was dealing with a very talented person. He started sending me portfolios of 50 stocks on a monthly basis that were the outcome of his dynamic models and I was stunned. These portfolios were going up consistently and sometimes even more than 15% in a month. We had many more meetings to get to know each other better and by the end of 2003, we back-tested the strategy in a global universe and in an Asian universe. The results, cleaned for survivor bias, were outstanding despite the adverse market circumstances in the past 11 years. That is the reason we decided to set up Quant Asset Management.
To fully concentrate on QAM, Chatchai left Thomson Financial and I parted with my stake and activities in my Asian fund.
I would say he is a true academic hobbyist and he is never happier than when behind the computer. Previously at Thomson, he would spend evenings and his weekends back-testing a huge amount of factors and factor combinations trying to find a way to beat the markets. He has read all the quant research that is out there but was never really impressed by what he saw. Often, even the most advanced quant shops would use multi-factor models in a very static way. He found the key to outperformance to be in dynamism as opposed to the so-often practised dogmatic approach like growth or value and bull or bear market dynamic in the sense that models must have their own artificial intelligence, able to listen to what the market tells them within a factor framework that makes both theoretical and practical sense.
When I first approached him with the proposal to set up a company together, he didn't like the idea too much. He was very happy at Thomson and had never thought of the monetary advantages of running a hedge fund. Only three months after I had first approached him did he come back to me with more enthusiasm to set up QAM.
As for me, I started my career at the same university (Utrecht, Holland) as where I obtained my master's in business law. I soon got bored with the university environment and got myself involved in the financial markets. I spent three years at ABN AMRO and later as a director at Merrill Lynch in London for seven years. At Merrill Lynch I not only learned a lot about how analysts do their work, I also learned how brokers make money. That is useful because there is a fine line between paying too much and getting not enough in return, and paying too little and not getting the service needed. I used to deal for big institutions and was involved in program trading. The way we execute our order flow is always through program trades, so my experience here is relevant. I left Merrill Lynch because I felt that being on the sell-side wasn't satisfactory. I wanted to execute my own ideas. In 2001 I co-founded an Asian equities long/short fund and that is where I got my experience as a hedge fund manager.
Our idea has always been that our company should be on the forefront of IT and we have developed systems and automated many processes within our company that many other companies still do manually. To illustrate this, we can update our investors on our NAVs automatically per email at every desired time interval. It is interesting to see that Singapore has recently overtaken the US as the world's best user of information and communications technology, according to a recently issued report called "Global Information Technology" by the World Economic Forum in Geneva.
Our office in Singapore is paperless. Automation is the reason why we don't need that many people to operate a global fund investing in more than 60 countries worldwide. The computers do the work. The only human input in our company obviously besides the skilful programming and execution is really constantly checking the systems. We currently also spend a lot of time on the processes of faultless execution, avoiding/minimising market impact and trading costs analysis.
First of all, the entire investment process is quantitative and is driven by the computer. Obviously our model is proprietary but in order to get investors interested in what we do and to make sure they understand our strategy we give away probably 60% of it. That is more than a conventional fund manager will ever give you, because he is simply not able to, due to an inherent lack of consistency in the human brain.
The idea that we run a black box model and are therefore not transparent and hard to follow is wrong. Just have a look at our newsletter and you will notice the level of transparency and consistency in our methodology.
Running the models takes many, many hours and involves millions of calculations, impossible for a human being or even a football stadium full of intelligent investment professionals to make. You have to realise that every month at the same date we run the models and the models will decide which factors need to be used and how they should be weighted. The models are let loose on a dynamic universe of 6,000 stocks for our Global Fund and 1,600 stocks for our Asian Fund. The factors used range from earnings forecasts changes, earnings revisions, earnings dispersion to price to book, price to cash earnings, etc. These are always divided into two main groups which we believe to be the drivers of share prices: earnings expectations and relative valuations.
We use the aggregate data from 50,000 analysts worldwide and if you think about it, earnings expectations always reflect every change in the relevant economic data be it at a company level or at a macro level. Interest rate changes, commodity price volatility, GDP growth expectations or what have you, it always comes back into analyst forecasts. Analysts are the first to take up these changes into their forecast because that is what they get paid for to do. Our models never look at the actual buy, sell or hold recommendations because that does not add any value.
Our portfolios always contain a big number of stocks from a minimum of 30 to as much as 120. It is a big numbers game with very low individual company risk. It is never a risky bet on the fortune or misfortune of a small number of companies.
Growth at a reasonable price and dynamism, ie, listening to the market, the place where the smartest people and institutions put their money to work. The stocks the models select always go through a valuation and earning growth/momentum filter. Sometimes weighted more towards valuation, sometimes weighted more towards momentum but never exclusively tilted to one side.
It is difficult to say because the models use many factors and strategies, and are dynamic. The strategy generally works well in bull or bear markets. Because part of what we do has an element of trendfollowing in certain factors, you could say that sudden trend changes can have a negative impact but due to the frequent rebalancing of the model, the model is able to quickly find the right track again.
Quite. All the processes are automated. The models select a huge number of stocks far more than a conventional fund manager would. The investment process is automated and entirely consistent and systematic. The issue we spend a lot of time on is liquidity. A position in a stock is never more than a certain percentage of the stocks' daily trading volume. That is what limits our capacity but we anticipate that we will be able to manage a total of US$800 million. We will only grow our funds after having shown step by step, year by year, that we can maintain our performance targets of an average of 40% per year.
It would be too extensive to sum them all up but basically any company worldwide is in our universe without restrictions as long as it is covered by at least 4-6 analysts and as long as it trades a minimum liquidity. The stock selection ranges and varies from emerging markets to developed markets and from small to big caps.
Maybe I would like to emphasise here that we really think we have a unique product, computer driven, programmed to generate alpha in either a bull or bear market environment. The analogy is in chess I guess. If you look at the history of it not long ago, it was consensus that a computer could never beat a human chess master. Look what happened. Now we all think that it is actually remarkable that human chess players even stand a chance against the computers we have today.
It is hard for human fund managers to consistently beat the market. It is easier on the inefficient side on the market than on the efficient side. In a universe where each stock is covered by let's say at least six analysts, the universe where we operate in, it is my conviction that you have to be systematic and computer driven otherwise you will not make it over a prolonged period of time.
Working with a value approach on the inefficient side of the markets where analyst coverage is low has proven to work as long as it is done by a talented fund manager. Concentrating on a universe of companies that are not well known to the investment community makes sense, and analysing them in-depth by a skilled human manager is the only other strategy besides our own systematic, computer driven strategy that I would contemplate investing in.
Yes, later this year or the beginning of next year we will launch the QAM Japanese Equities Fund and the QAM European Equities Fund based on the same methodology. These funds will have expected volatility and return profiles that are lower than our two current funds due to a different hedging methodology.
Contact Details
Frank Holle
Quant Asset Management Pte Ltd
+65 6549 7921
frank@quantasman.com
www.quantasman.com