Re:Gaussian vs. Mandelbrotian: The Great Intellectual Fraud
by mmt on Thursday September 10, 2009 @03:56PM
(#816049)
This makes sense, to be, so, perhaps, I'm confused by how he would then go on to talk about the individual values that make up the pair.
Is that, then, exactly his point? That a random sample of a group can't be modeled like a group of random samples, yet this is what's being done?
Given all pairs of people whose join income is $1M, select a pair at random. "All pairs" is an unusual population to select from, but mathematically it's a perfectly valid way to define a random variable.