CFE-CMStatistics 2024: Start Registration
View Submission - CFECMStatistics2024
A0966
Title: Addressing boundary inefficiency of local density estimators Authors:  Marco Di Marzio - University of Chieti-Pescara (Italy) [presenting]
Stefania Fensore - University of Chieti-Pescara (Italy)
Chiara Passamonti - University of Chieti-Pescara (Italy)
Abstract: Local density estimation methods suffer severe bias near the support boundaries due to the well-known estimated density overflow problem. Because of its practical occurrence, the counterpart of such phenomenon in regression assumes strong relevance, too. The so-called reflection method appears to be an efficient way to alleviate this issue. Simple domain transformations could favor an elegant application of such a principle.