Quantitative or 'quant' investing is an investment style in which mathematical or statistical models are used as inputs for investment decisions. Quant investing techniques are widely used by sophisticated professional investors, including a number of American and European hedge funds. the use of quant models as aids to supplement the judge of fund managers is growing and widely accepted as beneficial.
BUILDINGS QUANTITATIVE MODELS
Typically, quant models are implemented as computer software which monitors financial data, performs a set of calculations, and outputs investment decisions. for instance, a very simple statistical might input the closing values of the Nifty for the last 200 days, calculate the 200-day and 50-day moving averages, and output a 'buy' recommendation if the 50-day moving average has exceeded the 200-day moving average the past 7 days.
Prior to use in actual investing, the models are thoroughly back tested, using historical data to measure past performance. Back testing might flag defects in a model, or certain circumstances under which the model is less likely to make accurate predictions.
For instance, if a back test of the simple model described above showed that it lost money whenever RBI interest rate hike had occurred in the past week, the investor might decide to ignore or discount the model's output under suck circumstances in the future.
Over-reliance on the results of back tests can lead to seriously flawed model and may lead the investor to make the assumption that future market behaviour will continue to resemble past behaviour.
COMMON QUANTITATIVE INVESTING TECHNIQUES
There are some broad classes of techniques that are commonly used in quantitative investing.
Statistical Pattern Identification
Quant models will often crunch through large quantities of past data to identify hidden market relationships. One simple application often referred to statistical arbitrage is to find the prices of stocks which tend to move in the same (or opposite directions. For instance, while the stocks of oil exploration companies tend to benfit from a sharp increase in oil price, the airline stocks will suffer. A statistical model monitors oil prices, and trigger a call to go long the oil company and short the airline if oil prices have risen beyond a certain threshold.
Quantitative Value-Investing
Value investors buy stocks whi h seem undervalued relative to their intrinsic value. in theory, the intrinsic value of a stock can be calculated from a simple mathematical procedure called discounted cash flow valuation that discounts all future cash flows are highly unpredictable, and so the intrinsic value of a stock is very hard to accurately quantify.
Quant models seek to identify stocks that are cheap relative to fundamentals. for instance, they might rank all stocks in the ascending order of price-to-earnings ratios or price-to-book ratios. They might then buy stocks in the top 10% (low P/E ratio) and sell those in the bottom 10% (high P/E).
Momentum Investing
Research ha shown that under certain conditions, stocks that have out-performed in the recent past are likely to continue their out-performance in the near future. This phenomenon is called momentum or trend following, and is an important part of many quant trading models.
Momentum might occur because markets react slowly to positive news about a stock and the price will tend to increase slowly than spiking instantaneously. Also recent out-performance might cause a stock to gain investor attention and attract fresh buying interest. Quantitative momentum models typically identify out performers using metrics such as:
- Number of recent analyst upgrades or downgrades
- Percentage change in price over the past 3 months
Stocks that are out-performers are bought, and under-performers are sold.
PROS AND CONS
the use of quantitative models generates a great deal of debate of debate amongst the investment community.
Supporters of quant investing stress its advantages:
Objective: the use of a model reduces the need for sound real-time judgement on the part of an investor.
Speed: computer models can crunch enormous quantities of data, search for complex patterns and generate decisions at high-speed.
On the other hand, detractors criticize quant models as being
Simplitic: Markets are extremely complex, and many models make several simplifying assumptions, in effect ignoring that complexity.
Backward-looking: Models often rely on common, well-known relationships which have worked in the past, but need not continue to work in the future.
SUMMING UP
A number of sophisticated investors use quantitative techniques. Such techniques are growing in importance and popularity. Quant investing can be complementary, not necessarily a replacement for fundamental research and that it can function as an investment idea generator.
Read more...
