Customer Reviews

Confidence Intervals in Generalized Regression Models (Statistics: A Series of Textbooks and Monographs)

5 star:		(0)
4 star:		(1)
3 star:		(0)
2 star:		(0)
1 star:		(0)

 

 

This book is best read with a copy of the program, Mathematica, at hand. The approach is quite abstract, but with the examples you can do your own calculations. As the title indicates, the emphasis is on confidence intervals, as opposed to the older idea of a mean plus or minus one standard error. E.g., this older type of confidence interval for estimating a small probability could extend into negative numbers, a ridiculous result. I only wish the book could be extended to cover multivariate problems, such as 3-way contingency tables. Much of the programs in the CD attached to the book have been incorporated in the new version, 7.0, of Mathematica.