are likely to remain the subject of legitimate interest far into the future. In other words, they will continue to be useful in helping us understand not just financial markets but also human behaviour and psychology. An example of the many issues facing researchers is the shape of the security market line which, according to the capital asset pricing model, reflects the most efficient portfolio that would give the best return for a certain level of risk. As recounted by Campbell (2000, 1528-29), changes in the slope of this line led to several hypotheses that were attempts to discover the patterns of behaviour of such data. Amongst the conclusions that continue to influence equity markets is that of the significant contribution that small firms make to market returns. The digitalisation of data and the increasing power of number-crunching computer technology in the last twenty-five years certainly helped not only in gathering data and improving its integrity, but also in the development of mathematical models that somehow helped explain the data.
Second, many models have been developed in response to the data. ...
Coming in the form of equations (packed with Greek letters), these models help us to understand the reality that is captured by actual data. Several of the studies enumerated by Campbell (2000) helped in the development and our understanding of financial markets in the last twenty-five years. We are warned, however, on a point of caution implicit in the use of market data, especially as more academics attempt to find any observable patterns that are market anomalies (over-reaction and contrarian profit-making, month- or day-of-the-week effects, etc.). One key issue is the rationality (or lack thereof) of market investors and its connection to the integrity of the data. The continuing debate over the efficiency of capital markets between believers (Fama & French, 1998) and behaviourists (Shiller, 2000) call into question whether rational investor behaviour give rise to random data that irrational investors (as most investors are characterised by behaviourists) turn into predictable (and therefore, non-random) data through an act of rationality.
Third, the analysis and discussion of data and models have improved our understanding of the sources of risk, the economic forces that determine the rewards for bearing risk, and the factors that determine the over-all level of asset prices. The different asset pricing models have taught the investing public, businessmen, and public policy-makers several important and overwhelming lessons, such as: first, there is no such thing as a free lunch, and second, that the fluttering of a butterfly's wings in Argentina can bring down the management and the stock price of a company. What the first lesson teaches us is that the