Can I read Significance Testing in Correlation Analysis on EtoBox?
Significance Testing in Correlation Analysis by Saurav Ghoshal is a document available to read on EtoBox.
What is Significance Testing in Correlation Analysis about?
This document proposes testing a different null hypothesis when analyzing correlation coefficients. It argues that testing H0: μ = 0 is not optimal, as rejecting this null hypothesis does not provide much practical information unless the correlation coefficient explains a relevant amount of variance. The alternative proposed is to test H0: 0 < μ ≤ s, where s is the square root of the expected proportion of variance explained by linear regression. A simulation study found the actual type I error rate was clo
- Author
- Saurav Ghoshal
- Language
- EN