Re: Re: Re: Re: Re: Re: Re: Re: T'ai Chi : Some math/stat questions for you
BillHoyt said:
This is about the third demonstration from you that you don't understand basics about stat theory.
Hardly, dude. You can keep repeating that line of yours, but it still isn't true. Some advice: next time, ask me some harder questions, not ones like 'What is the t-test used for.' You ask me questions like then and then claim that my answers can be found anywhere. No ish Bill, maybe because you asked me incredibly simple questions. Ever think of that?
Chi-square's "Chi" is the sum of the squares of n standard Normals.
Right, gee, I wonder what I said. Oops, you forgot independent there, Billy. You might want to study more.
The net is a great resource once you understand the fundaments.
I agree.
Now the other area where you urped was in my twisted gaussian/binomial question. I figured you'd stop once you found the inversion. That is what you did, of course. The question wasn't about "approximations", though. It was about collecting n distributions together to get a different distribution. The answer revolves around the Central Limit Theorem.
Oh good god you've got it twisted.
First, you said, "CLM", when you obviosuly meant 'CLT'. Second, you asked how many normals do you need to get a binomial. It is the other way around. We approximate the binomial with the normal because, graphically, the binomial histogram fills out and becomes the normal curve.
Collect together a number of distributions of any sort, and you will eventually end up with the normal distribution.
I agree, 100%... Although you asked how many gaussians do you need to get a binomial. Do you know what distribution you get if you mix gaussians? Take a wild guess, Bill.
Oh, are you going to answer your own question, Billy? Tell us, how many gaussians do you need to get a binomial?
At least I didn't ask you about the two-sided Chi square test.
You should have. It would have been more difficult than that silly Student-t questions.
