Re: [Bug-gnubg] Ratings as random walks

[Joern Thyssen]

On Sun, Jun 15, 2003 at 04:44:16PM -0400, Douglas Zare wrote
> Quoting Joern Thyssen <address@hidden>:
> > 
> > Yes, but to get the 95% confidence interval you should devide by the
> > square root of the number of steps (following the discussion last week).
> 
> You are comparing apples with oranges. 

Yes, after some off-list discussion with Rod we came to the same
conclusion.

I have to dig out my books on statistics :-)

[snip]
> standard deviation should be higher. Since what you have observed is
> much lower, I don't think you have a representative sample.

Since the bot swings between 1900 and 2100 the std.dev. should be around
50?

Unfortunately I've no access to gammonvillage, so I can't follow the
link. Is the article available elsewhere?

Jørn