{"id":"arxiv/2602.14206","url":"https://arxiv.org/abs/2602.14206","tags":["math.ST"],"score":0,"title":"Kernel Estimation Of Chatterjee's Dependence Coefficient","domain":"math.ST","source":"arxiv","authors":["Mona Azadkia","Holger Dette"],"reviews":0,"subject":"math.ST","summary":"A dependence coefficient suggested by Chatterjee (2021) has attracted considerable attention in recent years. It takes values between 0 and 1, vanishes if and only if the corresponding random variables $X$ and $Y$ are independent, and equals 1 if and only if $Y$ is almost surely a measurable function of $X$. The coefficient coincides, for distributions with continuous marginals, with a dependence coefficient introduced in Dette, Siburg, and Stoimenov (2013). Several estimators of this measure have been proposed, including the kernel-type estimator of Dette, Siburg, and Stoimenov (2013),…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-02-15","reactions":{}}