Epistemology of Trends

2004-04-08T13:26:00.000-05:00

The Guardian Piece may help set the tone for the proposed event study surrounding the misdealings for which Martha Steward was recently convicted. Stay tuned for more on this topic from Laura Ng.

2004-04-01T12:05:00.000-05:00

Stock commentators are famous for talking about trends, but even the most rabid efficient market theorist can hardly avoid using analogous terms. Human beings have pattern recognition as part of their genetic heritage, and it is almost impossible for us not to see patterns --- even though the patterns we see are sometimes only artifacts of chance.

Consider the stock chart for MSO, Martha Stewart Living Omnimedia. Can you --- can anyone --- not believe that there was a several month down-trend as news of possible misdealings first gained press coverage? Wasn't there also a modest up-trend as the Martha Steward Trial began? Wasn't there a blow-up followed by a classic "dead cat" bounce after Ms. Steward was convicted?

All of these questions seem like simple common sense to anyone except those who are card-carrying members of the academic finance community. For those folks, these suggestions are no more sensible than seeing a "Man in the Moon" or a "Mermaid in the Fog." Oddly enough, statistical tools seem to be powerless to prove that the academics are wrong.

So here is a little research project. Take the time series for MSO, take the series for the options (this is harder --- there are several pannels to contend with), take a chronology of the legal and publicity events attending this drama, and bake the whole mess into a pie. I think that a story will emerge which will tell us a lot about the behavior of traded securities.

This is an event study of a special, yet amusing, kind. I don't yet have a good name for this type of study, but I'll work on one for the next entry.

2004-03-28T06:30:00.000-05:00

The discussion of "Trends" is one of the media's favorite topics, yet there is there are almost no investigations of the ways that beliefs about trends evolve. This blog will collect and comment on links which bump up against the epistemology of trends, or, more plainly, why we believe what we believe about trends.

Here are two questions that illustrate what this blog will examine.

Are some phenomena are more readily believed to be trends than others, even when the "facts" supporting the trend identification are comparable?

Are there "thought leaders" who can create a "trend" just by their own influence?