Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. It only takes a minute to sign up.
Sign up to join this community
I have heard in the past that it is 'dangerous' to run a correlation analysis on two percent changes. Such as a situation in which you are trying to make a link to a medication or other treatment and you are trying to say "The people who responded the most to the treatment, also had the most improvement in x".
Why is correlating two percent changes not appropriate?
Percent change is something that is sometimes useful for comparing means but is not appropriate to compute on individual subjects. Reasons are detailed n Chapter 134 of Biostatistics for Biomedical Research available from https://hbiostat.org/bbr .
