Distribution estimation without assuming normality

Akifumi Okuno

Presentation information

[J] Oral

M (Multidisciplinary and Interdisciplinary) » M-GI General Geosciences, Information Geosciences & Simulations

[M-GI29] Data-driven geosciences

Mon. May 26, 2025 9:00 AM - 10:30 AM 201A (International Conference Hall, Makuhari Messe)

convener:Kenta Ueki(Japan Agency for Marine-Earth Science and Technology), Shin-ichi Ito(The University of Tokyo), Keita Itano(Akita University), Masaoki Uno(Department of Earth and Planetary Science, Graduate School of Science, the University of Tokyo), Chairperson:Keita Itano(Akita University), Kenta Ueki(Japan Agency for Marine-Earth Science and Technology)

9:45 AM - 10:07 AM

[MGI29-04] Distribution estimation without assuming normality

★Invited Papers

*Akifumi Okuno¹ (1.The Institute of Statistical Mathematics)

Keywords:Statistics, Non-normality, Distribution estimation

While normality is often implicitly assumed in various statistical estimation problems, the actual distribution of data may be skewed, multimodal, or deviate to some extent from a normal distribution. In this talk, we will discuss the problem of distribution estimation in situations where normality cannot be assumed. In particular, we will introduce a proposed method that enables efficient non-normal distribution estimation even in the presence of outliers. We will also discuss its potential applications.