Random Rambling

Michael Crichton, RIP

Michael Crichton passed away on November 4th, 2008. Mr. Crichton, “the father of the techno-thriller,” is the author of 21 books including The Andromeda Strain, Congo, Jurassic Park, Timeline, The Lost World, Prey, and State of Fear, selling more than 150 million copies and translated into 36 languages, with twelve made into films. The recipient of an Emmy, a Peabody, and a Writer’s Guild of America award for the TV series, ER, he is a graduate of Harvard Medical School, and has been Visiting Lecturer in Anthropology at Cambridge University; Henry Russell Shaw Traveling Fellow; post-doctoral fellow at the Salk Institute for Biological Sciences; and Visiting Writer, Massachusetts Institute of Technology.

Michael Crichton’s latest best-seller, State of Fear, an indictment of Climate Change Fear Mongering, may in time be viewed as a landmark, both cautionary and prophetic. Is environmental debate today, including global warming, bio-technology, and other issues, based on science or politics? Are popular accounts of such issues rooted in science or phantom risks? Are government policies focusing on the trivial while ignoring the real, and in the process wasting limited resources, crippling human innovation to address true dangers, and inviting tyranny?

Statistical Illusions:

There is nothing scientific about Climate Models. Climate Models in many ways resemble econometrics and statistical modeling used by Keynesian and Monetarist economists. Models are statistical illusions. Economists who make prognostications about the future based on econometric models are often proven wrong, and when they are right, it is purely out of luck. That doesn't stop economists from continuously making new statistical predictions about the future.

There is no denying that certain trends in economy can be prescienlty predicted, but that always comes from understanding of the laws of economics and an understanding of human behavior. These factors are subjective and impossible to quantify.

Then there is the Heisenberg's uncertainity principle: Observation affects the observed - Heisenberg was speaking about subatomic particles. Mises spoke about the same certainity regarding human behavior. Once people get the wind of certain trends, crowding into that trend will make the trend null and void. Or if human beings get to know that they are being observed for certain purposes, they consciously or subconciously adjust their behavior.

Statistics is one of the most misused branch of mathematics. You can look at data from the financial markets correlate it to planetory positions and may be able to find that they follow some obscure patterns, but that doesn't make it a scientific truth. There are infact financial advisors who help you make investments based on astrology ;-)

Statisticians look at patterns in the past and they project it into the future, but often to their dismay, the future fail to resemble the past. According to these geniuses, if market has been going up, their models will predict that they will continue to go up. If globe has been warming, it will continue to heat up. Then much to their chagrin, future fails to resemble the past.

Black Swan is a term popularized by statistician finance professor Nassim Nicholas Taleb, who wrote a book by the same title - "Black Swan: Impact of the Highly Improbable." The term Black Swan would have been an oxymoron in the good old days a few centuries ago, because swans by definition were always white. Then the Europeans started colonizing australia in 1788, and they found black swans for the first time.

Once of the recent black swan event was the historical election of Mr. Barack Obama to the presidency of the United States of America. It was a statistical impossibility. If you sample the data, none of the previous 43 U.S Presidents were of African American ethnicity, which should make some statisticians conclude that probability of a black person getting elected as the Commander in Chief is 0. Yes, Mr. Obama overcame great odds, but it is impossible to overcome zero odds, and one would be crazy to even attempt. We know there are more factors affecting the electability of a person than just his/her race.

Another example that Mr. Taleb sites in his book is the life of a Turkey. Turkey grows up on a farm. It's life is rather easy - the farmer takes care of it, every morning it wakes up and there is enough food to eat, provided by the farmer of course. One day is hardly different from the other. Then, one fine morning in november, Turkey gets jolted from it's sleep and gets its head chopped off - Happy Thanks Giving.

Global Warming theory is nothing like Theory of Relativity or Boyle's law. There are very little if any, absolute, proven, and quantifiable relations between the various factors that affect the climate. Scientists may not even know entire list of factors that affect global climate and how precisely these factors interact with each other.

Is it really random rambling? or is there a statistical pattern to it?