OT--------------A Non-Random Walk Down Wall Street, the book.
I don't know if anyone read this book or not. It is a brand new book coming out of the university professors and hit the bookstores everywhere. The two authors are from MIT and Penn. I assume they are excellent and a lot of tricks we can get from them. After all, that is what many people go to school for.
The book itself is very math-intensive, and their conclusions are claimed in a very confused language, at least to me. I have only gone through first few pages and I can only draw very few points out of it. Hopefully someone also can help to explain more.
First, the authors are "number" professors, like many others. What I mean is that they have absolute no experience on Wall Street. Their research is purely based on "number". A lot of claims and speculations they made were funny. Only two Wall Street firms they mentioned might test their results. I can feel the resentment from one school of thoughts to against another among themselves.
Second, the references from the book were almost way back to 1980s, few were early 1990s. This means the authors either did not do more studies or there is nothing new anymore to study. Either way, this is a "old" research results.
Their main conclusion is that from the weekly data, the market is not "random walk", and might be random walk on the daily or the monthly basis from other research results. Basically it says there is a trend on the weekly basis (they probably could not explain it in a plain language). Let assume what they claimed is true. I suspect the non-random-walk cycle is much smaller now: not on weekly, but on daily basis. This is due to the birth of internet. People can access information much much faster and the commissions are very low. Human's short-term memory is getting shorter due to the arrival of the new information, i.e., the time to reach the maximum entropy is much faster.
I need to read more to figure more. So far nothing much new from the book. |
