{"id":"arxiv/2610.06873","url":"https://arxiv.org/abs/2610.06873","tags":["math.ST"],"score":0,"title":"Entropy-Based Goodness-of-Fit Tests for Generalized $κ$ -Gaussian Distributions","domain":"math.ST","source":"arxiv","authors":["Mehmet Siddik Cadirci","Konstantinos Zografos"],"reviews":0,"subject":"math.ST","summary":"We introduce this class of multivariate Kappa-Gaussian distributions by maximizing a type of Kaniadakis $κ$ entropy subject to certain first- and second-moment constraints. This differs slightly from the approach of Rényi and Tsallis, because the $κ$-entropy arises from a relativity-consistent deformation of the exponential and logarithmic functions and yields a family of distributions whose tails follow a power law, a family that actually reverts to the standard Gaussian distribution as $κ\\to 0$. We then create a $k$-nearest-neighbor (NN) estimator for $κ$-entropy and describe it using the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-09-08","reactions":{}}