Re: chi-square test goodness of fit
From: George Kahrimanis (anakreon_at_hol.gr)
Date: 11/05/04
- Next message: Osher Doctorow: "Multivariate PI LAD Regression"
- Previous message: Bruce Zweig: "Re: Why were the exit polls wrong?"
- In reply to: Alexis Gatt: "chi-square test goodness of fit"
- Messages sorted by: [ date ] [ thread ]
Date: 5 Nov 2004 14:00:05 -0800
"Alexis Gatt" <alexismajordomo@yahoo.co.uk> wrote in message
news:<cmd8or$b20$1@iss-nntp.leeds.ac.uk>...
> [...]
> According to a chi-square table I found in a book, the equivalent
> probability value is between 0.5 and 0.75.
> And this is where I am confused. Does this mean that the model
> predict the data well or not? And why?
I know the feeling. You had better struggle with this question
a couple of more days, but if you are in a hurry, read on now.
A goodness-of-fit score of about 50% means that your fit is of
average "goodness". A 100% goodness of fit means that the curve
goes through the points; a perfect fit, but people will suspect
that you have massaged the data *and* that you are obsessive-
compulsive. Consequently, not only it is nonsense to say that
100% g.o.f. is twice as good as a 50% g.o.f., but it may be less
desirable.
~ George Kahrimanis
- Next message: Osher Doctorow: "Multivariate PI LAD Regression"
- Previous message: Bruce Zweig: "Re: Why were the exit polls wrong?"
- In reply to: Alexis Gatt: "chi-square test goodness of fit"
- Messages sorted by: [ date ] [ thread ]
Relevant Pages
|
