31 Jul 2002 Bram   » (Master)

A Species of Gamblers

I finished reading Fooled by Randomness. What a great book. Taleb is a true man of science, and this book contains none of the usual financial bluster.

There are many interesting points made in this book. Highlights include -

  • If you look at stock markets globally historically, they aren't nearly as stable as the US stock market has been, and even in the US stock market the feature of never having fallen over a 20 year period is probably an accident of history. The risks of investing in the stock market are seriously underplayed.

  • Most successful traders are probably just plain lucky. Our view is distorted by forgetting about the ones who lost and were forced out of the industry.

There is a striking similarity between trading and programming in that both are so emotionally demanding that many, if not most people can never do them well. Good trading involves admitting that most of your performance is random. Of course, most successful investors will never accept that fact, and get very angry if you suggest it. Thankfully, the market has now crashed, so I no longer routinely offend people when the subject comes up in conversaton.

Good programming involves admitting your fallibility, with bugs and initially poor design being a fact of life. With only a little bit of cynicism, most programming Methodologies can be seen as very transparent attempts to not accept those facts. Their self-assuredness that such a thing is possible despite a total lack of evidence is actually quite childish.

All mathematical concepts in this book are explained very clearly in plain english, which is great for a populist book, but left me wanting more rigor. I may read Taleb's other book.

Reading this book made me very much want to get into finance. Every time my life starts to get in order I figure out some way to make it chaotic again, so I might as well have work which has chaos built in and pays commensurately. The software and mathematical problems involved are quite interesting, and there's even some social benefit justification - bad investment helps turn the markets into a gamble, good investment helps keep peoples's savings safe.

Investing in Angels

I have an interesting strategy for the angel in the angel problem.

The angel imagines herself doing a random walk of legal moves on non-eliminated squares. She keeps a record of what the path is, but truncates it down to nothing every time her walk returns to the original square. Over time, one of the available first moves will spend the largest fraction of time as the first step in her walk, and she makes that move.

This strategy does a good job of avoiding traps no matter how deceptively and contortedly they're shaped. How one might prove that it always works, however, I have no idea.

Latest blog entries     Older blog entries

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!