On 9 Dec 2003 at 10:01, Rich Ulrich wrote: One very good book discussing in detail analysis of Plackett-Burman designs and other irregular designs (irregular meaning that correlations between contrasts can be other than zero and 1), is C. F. Jeff Wu and M. Hamada: "Experiments: Planning, analysis, and Parameter Design Optimization", especially chapter 8.
But if you are only interested in estimating main effects, the analysis should be straightforward not requiring anything special. The Plackett-Burman design is an orthogonal array, so the estimates of main effects are uncorrelated. The special methods given in the above reference is for allowing these designs to be used for detecting interactions also. Kjetil Halvorsen > On 9 Dec 2003 04:40:43 -0800, [EMAIL PROTECTED] (F?rat �ZDEM?R) > wrote: > > > Hello, > > I am interested in analysis of plackett burman designs.There are > > many books brifly describing how a plackett burman design consructed > > and properties of its design matrix. But I haven't faced a book or > > web site explaining analysis of the design in detail. I especially > > want to learn calculation of sum of squares.( as I experienced it is > > different from any other fractional factorial design.) > > > I don't know what P-D designs are, are but google gives > hundreds of hits in English for > < "plackett burman designs" > > > Limiting the searches, I get 14 and 42 hits, respectively, > for my usual way of cutting back extra ones, > < "plackett burman designs" FAQ > > < "plackett burman designs" tutorial > > > - and 47 hits, which is more than I expected, for > < "plackett burman designs" "sums of squares" > > > I mention 'English' because I don't recognize the .tr domain, > and my newsreader shows me the poster's name with > two question marks for letters -- so I figure that his defaults > in google might be different. > > -- > Rich Ulrich, [EMAIL PROTECTED] > http://www.pitt.edu/~wpilib/index.html > "Taxes are the price we pay for civilization." > . > . > ================================================================= > Instructions for joining and leaving this list, remarks about the > problem of INAPPROPRIATE MESSAGES, and archives are available at: . > http://jse.stat.ncsu.edu/ . > ================================================================= . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
