- Series: Cambridge Series in Statistical and Probabilistic Mathematics (Book 25)
- Hardcover: 346 pages
- Publisher: Cambridge University Press; 1 edition (June 8, 2008)
- Language: English
- ISBN-10: 0521865069
- ISBN-13: 978-0521865067
- Product Dimensions: 7 x 0.8 x 10 inches
- Shipping Weight: 1.8 pounds (View shipping rates and policies)
- Average Customer Review: 5.0 out of 5 stars See all reviews (1 customer review)
- Amazon Best Sellers Rank: #5,247,007 in Books (See Top 100 in Books)
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Design of Comparative Experiments (Cambridge Series in Statistical and Probabilistic Mathematics) 1st Edition
Use the Amazon App to scan ISBNs and compare prices.
Featured Springer resources in mathematics
Explore these featured titles in mathematics. Learn more
Customers who viewed this item also viewed
"... a carefully planned, well-organized [manner]. Using very simple language, Bailey makes every effort to make this book enjoyable to read, and to bring appreciation to and a deep understanding of the concepts and issues in experimental designs."
D.V. Chopra, Wichita State University for Choice Magazine
For every practising statistician who designs experiments, a coherent framework for the thinking behind good design. Also ideal for advanced undergraduate and beginning graduate courses. Examples, exercises and discussion questions are drawn from a wide range of real applications: from drug development, to agriculture, to manufacturing.
Browse award-winning titles. See more
If you are a seller for this product, would you like to suggest updates through seller support?
Top Customer Reviews
I once watched a sculptor create a statue. I was curious about the order in which she selected her tools. Knowing that this was an avocation, and that she also was an expert in robotics, I asked her about her selection criterion. She laughed at me (knowing I was in the same field) and said: "We scientists are always looking for the inductive abstraction, the process we can generalize. But in true art there is no SINGLE algorithmic order of tool selection you can assign to a factory floor. The emergent reality, in the moment, interacts with the artist, and the tool is selected based on much more exacting need. This is true whether you're a sculptor working in clay or a digital artist working in Maya."
That little digression aptly describes this text! In a perfectly ordered fashion, it manages to detail both the art and science of matching statistical tools to experimental design. The way Dr. Bailey does this, however, is amazingly applicable to algorithmic implementation! The razor's edge of real examples, problems with each design, work arounds and very unique ways to look at matrices (latin squares have no better treatment in any text, anywhere) balances both the art of tool selection and the generalization needed as a new kind of "design pattern" so we're not continually reinventing the wheel either.
This concept might already seem strange to mathematicians and programmers, but I know of one field where it is still VERY true: Organic Chemistry. The folks there still respect long experience and a "nose" for molecular arrangement that only comes after years of design, and still is as much art as science. If you've read the Black Swan, you know that we stats types often take way too much for granted, and Rosemary is well aware of the pitfalls of that frame in every chapter. Being honest with each other, we know that the "back end" of many studies are so carefully analyzed that we often gloss over both what was selected and why (opportunity costs) and the foundation design.
Background needed includes linear algebra, at least two undergrad years of stats including of course regression and ANOVA. A third year (undergrad) course in more advanced LA including the eigenspaces of symmetric matrices and orthogonal projections. The author doesn't say this, but I'd add a good grounding in polynomial operators if you want to extend what you're reading to algorithms and especially R.
Please note that although this is both a text and a reference, it gives detail in DESIGN, but not detailed R implementation. By design I mean drawing the diagrams, choosing the blocks and assigning the relationships. Execution of the stats can happen with any package you like-- this (in a sense) is about something both far more basic and far more important-- avoiding foundational mistakes BEFORE you crunch the numbers!
Highly recommended. Now-- for those of you that are autodidacts like me, without a big company or university to buy this for us: it is quite expensive now due to high demand and low supply. Cambridge is more academic, and I don't think they anticipated the uptake of this volume not only for training but for reference! They should have believed their own author. Solution: Instead of paying $200 US, check out both Amazon's third party sellers and their sister company at Abe Books. I got this for $30 US nearly new. You can do it, but you need a little patience. Either way, highly recommended!
EMAILER KINDLE ANSWER: I love my Kindle Fire, but NO, this is absolutely NOT recommended for Kindle (or any other E-reader). It is FILLED with diagrams, LaTex, MathML, <math> formula </math> tags, illustrations, etc. which are slaughtered by e-readers at this stage of the technology. Even MathType and other conversions can't help it because Rosemary's diagrams are key and intricate. Besides, for the $35 US or so for Kindle at this writing, you can get the paperback.