The Large Sample Behavior of Transformations to Normality.
WISCONSIN UNIV-MADISON DEPT OF STATISTICS
Pagination or Media Count:
We investigate the large sample behavior of both the classical and Bayesian procedures for selecting a transformation to normality. The study of the large sample behavior clearly reveals the role played by the assumptions leading to the Box and Cox procedures. Based on our large sample results, we introduce an information number approach for transforming a known distribution to near normality. This latter procedure provides bench marks for the maximum possible amount of improvement through power transformations. We illustrate our procedure with three examples. Finally, we generalize our procedure to random vectors and linear models situations. Author
- Statistics and Probability