Fooled by Randomness by Nassim Nicholas Taleb
From Taleb's mouth himself: "[Nassim Nicholas Taleb] is only interested in one single topic, chance (particularly extreme & rare events); but it falls at the intersection of philosophy/epistemology (skepticism; knowledge about the dynamics of history; inferential claims), philosophy/ethics (stoicism facing random events; theories of nonhedonic happiness), mathematics (probability, statistical physics), social science/finance (opacity & incomplete information; why economists have no clue but think that they know a lot), and cognitive science (how we are “fooled” by randomness). He mainly derives his intuitions from a 2-decade long and intense practice of derivatives trading (“nondull” activities with plenty of randomness). "
This book really is about many topics (brought together by the common thread of randomness), though my favourite points were about skepticism. So much of what we do is defined by what we know, that we often ignore or fail to accurately represent what we don't know. And what we don't know can bite us in the ass. Specifically if the outlier events are the radical large events that are probabilistically small yet large enough to skew the average. Taleb takes David Hume's definition of a black swan, and runs with it: "No amount of observations of white swans can allow the inference that all swans are white, but the observation of a single black swan is sufficient to refute that conclusion" (up until the discovery of Australia and their black swans, the English assumed all swans were white).
Overall, my review of this has to be funny. I will strongly recommend that everyone read this book. Fooled by Randomness has a great examination of probability, a wonderful semi-philosophical examination of the weaknesses of our human mind, and an especially fun writing style and clever examples. But I can't call it a great book. The author often seems arrogant, and his own skepticism leaves you with doubts about his examples. Other people (in the reviews at amazon.com) have problems with his methods. If nothing else, I've deeply enjoyed it, and will definitely read this again.
Also, if you like the book (or would like to know more), Nassim Nicholas Taleb has a lecture at PopTech! 2005 that is a good summary of most of his ideas. The Scandal of Prediction. (previously blogged by me here).
This book really is about many topics (brought together by the common thread of randomness), though my favourite points were about skepticism. So much of what we do is defined by what we know, that we often ignore or fail to accurately represent what we don't know. And what we don't know can bite us in the ass. Specifically if the outlier events are the radical large events that are probabilistically small yet large enough to skew the average. Taleb takes David Hume's definition of a black swan, and runs with it: "No amount of observations of white swans can allow the inference that all swans are white, but the observation of a single black swan is sufficient to refute that conclusion" (up until the discovery of Australia and their black swans, the English assumed all swans were white).
Overall, my review of this has to be funny. I will strongly recommend that everyone read this book. Fooled by Randomness has a great examination of probability, a wonderful semi-philosophical examination of the weaknesses of our human mind, and an especially fun writing style and clever examples. But I can't call it a great book. The author often seems arrogant, and his own skepticism leaves you with doubts about his examples. Other people (in the reviews at amazon.com) have problems with his methods. If nothing else, I've deeply enjoyed it, and will definitely read this again.
Also, if you like the book (or would like to know more), Nassim Nicholas Taleb has a lecture at PopTech! 2005 that is a good summary of most of his ideas. The Scandal of Prediction. (previously blogged by me here).