New: Book Report: Introduction to the Practice of Statistics

Today, I'm playtesting the Hogwarts Game. This game is going to be kind of unusual in that Game Control is providing transportation. They warned us not to pack our stepladders, copy machines, coolers, or any other heavy things--because we'd need to carry them all, and we wouldn't always have a van to stow them in.

This begged the question: how much can I carry in a backpack while walking around without getting winded? Can I carry a laptop--and other things at the same time? And so I took some walks around San Francisco carrying a laptop and the book I was reading at the time--a big statistics textbook.

Yes, I read a statistics textbook. At work I find myself dealing with big piles of data. I want tools to help me to understand big piles of data. The statistics that I learned back in high school helps, but I could use more.

The main thing I got from reading Introduction to the Practice of Statistics is the Central Limit Theorem:

The sampling distribution of (the mean of x) is normal if the underlying population itself has a normal distribution. But what happens when the population distribution is not normal? It turns out that as the sample size increases, the distribution of (the mean of x) becomes closer to a normal distribution. This is true no matter what the population distribution may be, as long as the population has a finite standard deviation.

Everyone who's studied statistics is probably rolling their eyes that I pointed out such a basic thing. But trust me, if you haven't been allowing yourself to use many tools because they were only good for normal distributions, and you were stuck, then this is a big deal. Go ahead, roll your eyes at the slow-paced self-taught bozo. I'm happy.

Tags:  |  |  |

Labels: ,

Posted 2006-08-26

 Avani Wildani said...

I still think the central limit theorem is one of the CoolestThingsEver (tm). I remember being really excited when I figured out how they made the tables in the back of the book :-p

26 August, 2006 13:19