# Validation of Association

**Authors:** Ćmiel Bogdan, Ledwina Teresa  
**arXiv:** [1904.06519](https://arxiv.org/abs/1904.06519)  
**Date:** 2019-04-13  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/1904.06519v1/

## Key points

- Authors discuss the crucial task of recognizing, quantifying, and visualizing associations between two variables
- Introduce a novel function-valued measure of dependence known as the quantile dependence function
- Quantile dependence function plays a pivotal role in constructing tests for independence and allows for easily interpretable diagnostic plots
- Specifically designed to identify general dependence structures between variables within different quantiles of their joint distribution
- Analyzing these structures across various parts of the distribution provides valuable insights into underlying relationships between variables
- Develop new estimators for the dependence function and explore their properties
- Utilize them to devise innovative tests for independence
- Demonstrate efficacy of proposed tests through numerical experiments compared to established independence tests from previous studies
- Showcase real-data analyses illustrating application of tests to uncover and visually represent complex dependence structures
- Provides a comprehensive framework for assessing and validating associations between variables using advanced statistical methods and graphical tools

## Summary

In the paper "Validation of Association" by Ćmiel Bogdan and Ledwina Teresa, the authors discuss the crucial task of recognizing, quantifying, and visualizing associations between two variables. They introduce a novel function-valued measure of dependence known as the quantile dependence function. This measure plays a pivotal role in constructing tests for independence and allows for easily interpretable diagnostic plots that highlight deviations from the null model. The quantile dependence function is specifically designed to identify general dependence structures between variables within different quantiles of their joint distribution. By analyzing these structures across various parts of the distribution, researchers gain valuable insights into the underlying relationships between the variables under study. The authors develop new estimators for the dependence function and explore their properties. They also utilize them to devise innovative tests for independence. Through numerical experiments, they demonstrate the efficacy of their proposed tests compared to three established independence tests from previous studies. Additionally, they showcase real-data analyses that illustrate how these tests can be applied in practice to uncover and visually represent complex dependence structures. Overall, this paper provides a comprehensive framework for assessing and validating associations between variables using advanced statistical methods and graphical tools.