0

Is it possible to do a variation of a root data transformation that is below x^(1/2)? Are there rules against this - for example: do they need to be whole numbers? Can anyone provide literature on this?

Context of transformation: Transforming to exhibit a normal distribution before applying a specific statistical test.

Example: x^(1/1.5)

Danib90
  • 407
  • 2
  • 5
  • 12
  • https://stats.stackexchange.com/search?q=box+cox+-regression+score%3A2 – whuber Dec 11 '18 at 19:21
  • Re the context: you might want to read https://stats.stackexchange.com/questions/25738/is-a-log-transformation-a-valid-technique-for-t-testing-non-normal-data/25753#25753 first. – whuber Dec 11 '18 at 19:25

0 Answers0