论文信息 - TRANSFORMATIONS OF MULTIVARIATE DATA

TRANSFORMATIONS OF MULTIVARIATE DATA

SUMMARY Methods, which are extensions of the techniques of Box and Cox [19641, are proposed for obtaining data-based transformations of multivariate observations to enhance the normality of their distribution and also possibly to simplify the model (e.g. improve additivity, homoscedasticity, etc.). Specifically, power transformations of the original variables are estimated to effect both marginal and joint normality. A method for improving directional normality is also described. Examples are included to illustrate some properties of the methods.

J. L. Warner | D. F. Andrews | R. Gnanadesikan | D. Andrews

[1] Bartlett Ms. The use of transformations. , 1947 .

[2] F. J. Anscombe,et al. THE TRANSFORMATION OF POISSON, BINOMIAL AND NEGATIVE-BINOMIAL DATA , 1948 .

[3] J. Tukey,et al. Dyadic anova, an analysis of variance for vectors. , 1949, Human biology.

[4] C. Daniel,et al. A Multifactor Experiment , 1954 .

[5] J. Tukey. On the Comparative Anatomy of Transformations , 1957 .

[6] James L. Dolby,et al. A Quick Method for Choosing a Transformation , 1963 .

[7] D. Cox,et al. An Analysis of Transformations , 1964 .

[8] R. Gnanadesikan,et al. Graphical Methods for Internal Comparisons in Multiresponse Experiments , 1964 .

[9] D. Fraser. Data Transformations and the Linear Model , 1967 .

[10] William G. Hunter,et al. Transformations: Some Examples Revisited , 1969 .

[11] N. E. Day. Estimating the components of a mixture of normal distributions , 1969 .

[12] David R. Cox. The analysis of binary data , 1970 .

[13] J. N. Srivastava,et al. Analysis and design of certain quantitative multiresponse experiments , 1971 .

[14] D. F. Andrews,et al. A note on the selection of data transformations , 1971 .