Mathematics
3 years ago

New nonparametric test revolutionizes linear regression analysis for small samples!

Journal of the American Statistical Association
A Nonparametric Test for Linear Regression Based on Combining Kendall's Tau With the Sign Test
J. F. Lancaster, Dana Quade

Paper Summary

Related papers

Theme background image
Mathematics
3 years ago

New nonparametric test revolutionizes linear regression analysis for small samples!

Multicollinearity
1 view
Theme background image
Computer Science
3 years ago

New test improves accuracy of high-dimensional data analysis in research.

Model-Based Clustering
Skew Distributions
Missing Data
3 views
Decision Sciences
5 years ago

Choosing the Right Test Statistic: Key to Accurate Hypothesis Testing

Research Methods Business
Statistical Literacy
2 views

A new test for linear regression parameters was developed using the signs of residuals and their correlation with the independent variable. The test's distribution in small samples was calculated, and approximations for larger samples were derived. The test's efficiency was compared to traditional least squares and Brown-Mood tests. Key findings include the creation of a confidence region and the calculation of Pitman efficiency.