It is ironic that the term quant" is used to describe something that is perceived to be very precise, yet its usage is anything but precise! Indeed it is widely used to cover a broad range of investment approaches that really have one thing in common: they are seen as being different in methodology from qualitative or traditional approaches.

The label "quant" can be applied to anything from passive index funds or tilted index funds to full valuation and optimisation techniques, tactical asset allocation and arbitrage strategies that use derivatives. However, a few common characteristics go some way to justifying putting all such techniques under the same label. For example, all rely on quantified data as the basis for decision-making, all have a defined quantitative benchmark to beat and use a computer program or programs. But at a fundamental level there are significant differences. The objectives can be very different, from matching an index return to outperforming an index to maximising total returns. The underlying processes can also be different. Using stratified sampling techniques to build an index fund, for instance, has little in common with the valuation techniques used to price option contracts.

It is possible to suggest that the term is so imprecise and covers such a disparate group of strategies that it contributes little to overall understanding. Yet there is in fact a common core of quantitative approaches which do have sufficient aspects of commonality to warrant consideration as a meaningful group that can be contrasted with qualitative approaches.

The contrast is particularly clear in the subset of active equity managers. Qualitative or traditional managers pride themselves on their ability to discern unique features of an individual case - potential that others have not (yet) recognised. The strength of quantitative analysis is systematic comparison of common characteristics across a broad universe, using models developed to identify securities that will outperform. In other words, quantitative investment relies on rigorous evaluation of comparable alternatives, while qualitative investment depends on anecdotal analysis.

Another way of considering the differences is to think in terms of the application of scientific tools of inference as opposed to the more associative and intuitive or artistic. By definition, the size and nature of the universe is the key. Quant tools are applied systematically to a large number of securities, all of which are given equal consideration. For example, at Rosenberg our models consistently value over 12,000 stocks globally. Traditional analysts winnow down the universe so that more and more is known about fewer and fewer stocks. The data the quant investor uses is primarily publicly available, but the insights are proprietary. The traditional manager starts with publicly available data but supplements it with proprietary information, for example gleaned from visiting a subset of companies.

Highlighting these differences may be instructive in itself, but it is really the implications for the investment product that are of most interest. Qualitative managers are clearly susceptible to bias, since intuitive judgments are the core of their strength. As a result there is a tendency to make significant "bets", which will undoubtedly result in volatile performance. Quants, by contrast, tend to identify and exploit smaller mispricings, giving a somewhat smoother, but less "exciting" ride.

In other words, the answer to the question posed in the title is: quant is something investors who want to sleep at night!

Jennie Paterson is managing director of Barr Rosenberg European Management"