Bayesian variance-stabilizing kernel density estimation using conjugate prior (Q2314517)

It is noticed that kernel-type density or regression estimator does not produce a constant estimator variance over the domain. Several authors proposed new methods to correct this problem. A variance-stabilizing (VS) local variable bandwidth for kernel estimators are used by the author and \textit{Y. Kanazawa} [Bull. Inf. Cybern. 43, 53--66 (2011; Zbl 06064008); Commun. Stat., Theory Methods 44, No. 10, 2151--2175 (2015; Zbl 1329.62173)]. Convex combination of three skewing estimators is used by the author [``Skewing methods for variance-stabilizing local linear regression estimation'', Preprint, \url{arXiv:1704.04356}]. In new approach to stabilize the variance a Bayesian-type of variance-stabilizing is used.