Yes, that was an excellent book. I posted about it a few times:

http://digitaltonto.com/2012/the-problem-with-pundits/

http://digitaltonto.com/2012/when-should-we-go-with-our-gut-and-when-should-we-look-before-we-leap/

http://digitaltonto.com/2012/irrational-expectations/

– Greg

]]>Thanks for your input Emily.

– Greg

]]>Emily,

Yes. I also believe that outliers have been overlooked. As I said in the post, they’re the data points that are really interesting.

However, I think what is probably the more important issue is when outliers aren’t really outliers at all. Many people (most disastrously the financial community) assume a random, normal distribution when there is no reason to do so and they therefore undercount extreme values.

Since the financial crises, there’s been much more attention paid to “fat tailed” models but, as Nassim Taleb points out, even that doesn’t fully insulate you.

– Greg

]]>Thanks for sharing your input.

– Greg

]]>That’s actually a subject I’ll be posting quite a bit about over the next few months. Computers are getting scary good at pattern recognition and, if you believe Ray Kurzweil, our advantage will disappear by 2030.

It is a very big problem and it’s coming fast.

– Greg

]]>I also love Math and it represents an important part in my profession.

Yet I also think that the capacity of the human brain to deal with complex problems is unrivaled.

So, computers can help us to make fast calculations and iterations, but a mathematical model has to be at the same time simple enough to permit to us “problem solvers” to extract (and to abstract) from the numbers the patterns in order to have a better understanding of the problem to solve.

]]>