The Interaction Algorithm and Practical Fourier Analysis
Irving John Good · Journal of the Royal Statistical Society Series B (Statistical Methodology) · 1958
Summary The interactions in a complete factorial experiment are linear functions of the observations. They can be calculated by means of a stream-lined algorithm, of which the first example was the adding-and-subtracting algorithm of Yates (1937). The relationship between the two methods of calculating the interactions is given a precise and succinct form, which surprisingly does not seem to have been given in the existing literature. The algorithm is logically simplest for the tn experiment. The inverse algorithm is specified and also the method of obtaining the divisors for the analysis of variance. Finally some analogous short cuts in practical Fourier analysis are described.
Cited by 509 works. doi:10.1111/j.2517-6161.1958.tb00300.x
Authors
References
- The Calculus of Observations.
- An Introduction To Abstract Harmonic Analysis
- A new finite series for Legendre polynomials
- The Design and Analysis of Experiments
- Die Mathematischen Hilfsmittel des Physikers
- “Design and analysis of industrial experiments”*
- Die Mathematischen Hilfsmittel des Physikers
- Random motion on a finite Abelian group