Nassim Taleb: The Black Swan: The Impact of the Highly Improbable
Ok, before I attempt a review, let me just say, this book is fucking bollocks, and another case of me being happy that I only paid the cost of postage.
Taleb's thesis is that investors, or more specifically 'quants' put too much stock in their models and fail to recognise the truly stochastic (random).
He points to things like regression models that assume that residuals follow a normal distribution (if they don't it's often standard practice to fit dummy variable in to essentially remove an outlier or outliers), and other models such as the Black-Scholes-Merton model that blew up in the faces of the academics that created it, as the crisis in Asia in the early/mid 90s saw their Long Term Capital Asset Management fund sink and require the government to bail them out.
Yes, this thesis is interesting but hardly new, people have know this for decades. Anyway, Taleb adds nothing new to this, rather he just provides a hundred different ways of re-stating the same thing. As well as this he continually alludes to how rich and successful he is, and some psycho-babble involving some fictional characters called Yevgenia and Nero Tulip.
This book was 'highly' recommended by the NY times and the Financial Times.
Really, it's a piece of pseudo-intellectual nonsense that offers absolutely nothing to anyone's understanding of finance or statistical modelling. It's simply one of the worst books that I've ever read.
Here he is on Newsnight:
[youtube=ABXPICWjFIo]
Nothing but truisms and no substance, very much like his book.