Association measures are important in quantifying the relationships between random variables in applications. In this work, we investigate the connection between Kendall’s tau and Spearman’s rho for zero-inflated data. First, we recall the state-of-the-art techniques to estimate \(\tau \) and \(\rho _S\) when the data exhibits a large number of ties at zero. Then we analyze how the values of \(\tau \) and \(\rho _S\) differ by focusing on a case study on the use of public transport in the Netherlands. We conclude the paper with a discussion of future interesting research directions on the topic.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Association Measures for Zero-Inflated Data with Application to Public Transport

  • Jasper Arends,
  • Elisa Perrone

摘要

Association measures are important in quantifying the relationships between random variables in applications. In this work, we investigate the connection between Kendall’s tau and Spearman’s rho for zero-inflated data. First, we recall the state-of-the-art techniques to estimate \(\tau \) and \(\rho _S\) when the data exhibits a large number of ties at zero. Then we analyze how the values of \(\tau \) and \(\rho _S\) differ by focusing on a case study on the use of public transport in the Netherlands. We conclude the paper with a discussion of future interesting research directions on the topic.