Customer Reviews

Experiments with Mixtures: Designs, Models, and the Analysis of Mixture Data (Wiley Series in Probability and Statistics)

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

 

 

The topic of this book is the design and analysis of experiments with mixtures of ingredients. This can be applicable to soft drink manufacturers, pharmaceutical development and many other problems in industry where different chemical combinations are used to produce the best products for a given purpose.

In statistics we commonly refer to mixtures of distributions that are used to model multimodal data. So statistician use to this terminology might think that is what the term mixture means here but it is not! Mixture strictly refers to the products under study being mixtures of chemical ingredients.

John Cornell is a professor of statistics who is acknowledged as one of the leading experts on mixture experiments. This book is his latest work and provides a clear description of the problem and its solutions. Anyone who works on experimenting with mixtures should have this book for a reference.

I'm a chemist. An old chemist. A very old chemist. One that has been doing mixture experiments since the mid-80's. Who apparently loaned out his Cornell never to see it again. I know of no other mixtures text book that handles the somewhat complex subject with this level of detail and clarity. Of course these days any number of software publishers offer mixture capability (I happen to use Minitab) which may make a text like this a bit less necessary. I bought this for two reasons - to refresh my fading memory on the matrix algebra details and to fill the empty spot in my book cabinet. I was pleased to be able to find this for fear it was out of print.