Site Meter

Saturday, March 27, 2004

Don't you get tired of being right all the time mr Clarke part XXIV

I agree with some articles and posts and such, which note that everyone makes mistakes., and so we should not decide Bush is unfit for office because with 20-20 hindsight we can see that there are things that the Bush administration could have done to prevent 9/11. I agree that Bush is unfit for office because he refused to learn from his (inevitable) mistakes. On 9/1/01 Bush was as arguably fit to be president as he was on 1/20/01. On 9/12 he lost that arguable fitness by obsessively insisting that a link between the guy who tried to kill his dad and the guys who killled a lot of peoples dads and moms Must be found, whether it existed or not.

While we should demand that the president learn from his mistakes and should vote Bush outin order to get a president who is modest enough to grasp that he can make mistakes and intelligent enough to learn from them,
it is unreasonable to expect that people never make mistakes when trying to predict the future.

For example Richard Clarke has made many mistakes. For example when he ....
Oh hell he hasn't been subject to muck scrutiny, he must have been wrong about something. No one is right all of the time.
Let's just take it as given that Richard Clarke has made mistakes. I'm sure he pooped his diapers once when he was a new born.

The question is how many people have ever batted 1000 ? OK so Richard Clarke that's one and ...
hey wait what is this on my bookshelf ?

Predicting the Signs of Forecast Errors

Robert Waldmann
EUI WP 95/22
May 3 1995

Abstract: It is possible to predict the signs of forecast errors using the difference between individuals' forecasts and the average of earlier forecasts of the same variable. Two simple methods for deciding which forecasts are far from the lagged average give the same result. Every forecast which is far above the lagged average is too high and every forecast which is far below the lagged average is too low. By combinding these methods, it is possible to improve 115 forecasts without worsening any. ....

Hey I managed 115 out of 115 in 1995 !

But wait, my claim was that whenever some guy says something waaay out of line with the conventional wisdom he is wrong and he should move closer to the conventional wisdom.

What about Clarke. Well on 9/4/01 he said "hundreds" of dead and we see that, like the rest of them, he underestimated the risk since by 9/11 there were thousands of dead.

That means I got 115 out of 116 right.

ohhhhh Nooooooooo. I am not batting 1000. The only person who has hit 1000 is Richard Clarke.


Referees for the International Journal of Forecasting take notice when I finally get around to resubmitting the manuscript.

No comments: