Quant investing? What's that?

Investors are sighing with relief that the markets are up again after a frightening plunge. Those who have been in the market from the early days of the bull run are happy to have sat through the trough.

One class of investors who benefited from the fall are mutual fund SIP investors. They picked up the lows by default and averaged their costs down. Implementing a discipline in investing through mechanical tools is as old as investing itself.

Its merits shine through during periods of volatility, when decision-making becomes tough. The realm of quant-based investing extends beyond the SIP and is growing rapidly.

Fund houses, hedge funds and institutional investors have taken the application of quantitative models in investment decision making to a new high. Quant models use a variety of techniques, such as fuzzy logic, neural networks, genetic algorithms, Markov models, fractal methods, and clustering techniques. The investment techniques they use draw more from physics than from economics.

It is estimated that three out of 10 hedge funds are purely quantitative model-based funds. Barclays Global Investors, the world's largest money manager, is a pure quant investor with $1.6 trillion in quantitatively managed funds. This category of funds has grown at twice the rate of the world's mutual funds in the last year.

The Times magazine recently carried a story which found that large cap funds run by quants consistently beat those run by non-quants since the beginning of 2003, by up to 2 percentage points a year.

Quants take about half as much risk as non-quants and do not lose as much money in the down years, it noted. It is now widely believed that in contrast with the individual style-based investing that was dominant in the 1990s, going forward, sophisticated quantitative tools that take the emotion out of investment decisions are likely to rule.

It is not as if quantitative investing has not been criticised. Many argue that selecting stocks is more an art than a science, not amenable to being captured in computer programs.

Quant models use extensive back testing of past data to create their investment algorithms, raising the issue that the past may not accurately represent the future. Some of the early techniques that used simple technical rules based on past price behaviour have been accused of being exercises in 'torturing data until it confesses'.

Then there is the danger that models would simply be replicated once they succeed and, therefore, follow a path of self-destruction. Sophisticated tools that used quant models to identify arbitrage opportunities not visible to the human eye have suffered such a fate in the past.

However, the cycle seems to be turning, driven by three factors, in favour of the quants.

First, the huge growth in the size of markets has made it possible to invest in these models and still get benefits of scale.

The Indian derivative market has grown in volume from zero to Rs 20,000 crore (Rs 200 billion) a day in less than 10 years, and is dominated by mechanical arbitrage strategies.

Second, the availability of computing prowess at low costs and the penetration of Internet trading have moved quant investing from the realms of professional managers to retail investors.

Several geeks are creating programs that enable them to make quick bucks from stock trading.

Some have extended these programs to pick stocks. Third, the investment industry has formalised quant investing. It is no longer taboo to tell investors that quant models are doing most of the work. For those who like machines over men when it comes to managing money, a new world of opportunities beckons.