Never be afraid to get dirty, but be sufficiently sure-footed to avoid the abyss of contamination.


Time for a little Q&A.

Dear Krish,

How can I be an expert like you?


Thanks for the question, Krrish! You flatter me. While various honorary degree offers flood my spam folder on a daily basis, that hardly qualifies me as an expert. What makes an expert? From what I’ve seen on television, to be an expert one has to be invited by Wolf Blitzer or some other news show host and make predictions as if they were irrefutable facts.

What if the expert is wrong? The beauty of being an expert is that it doesn’t matter. There are enough experts out there predicting different outcomes that one of them is bound to be right. And the one that is won’t let you forget it because that legitimizes the title he’s given on Wolf Blitzer’s show. He can also refer to the time he was right when one of his other predictions goes wrong. The world’s Fareed Zakarias, Howard Finemans, Robert Novaks, and William Kristols don’t all have to be right all of the time. Only one of them does in a given instance before you hear, “I told you so!” And it’s that wonderful phrase that makes this group of people experts. One of them correctly guessed what would happen, so all of them will be called back by Wolf Blitzer, Chris Matthews, and even Jon Stewart for the next prediction that just one (perhaps a different one) has to guess correctly.

If different experts spit out different answers, and we won’t know who is right without the benefit of hindsight, how does one distinguish intelligent decision making from random guesses? One way to judge an intelligent decision making strategy is to choose one that minimizes the amount of regret when compared to the best decision among the experts picked with the benefit of hindsight. In certain cases the notion of regret can be made mathematically precise and can then be used to model problems in universal prediction, machine learning, and game theory. One of the most notable uses of this model has been by Thomas Cover, who used it to show that the universal portfolio, an investment strategy he proposes, does almost as well as the best constant rebalanced portfolio subject to some assumptions. In this case, regret is measured in dollars and his experts were the ones choosing different constant rebalanced portfolios.

If one could make an intelligent decision without the help of an expert, would we still want them on the television? Well, who else is going to be there to say, “I told you so!”?


2 responses

  1. Sandhya

    i saw a movie named krrish. now THAT guy was an expert.

    January 29, 2007 at 5:32 pm

  2. Pingback: Past Posts « Dirty Hands

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s