About
Ph.D. in Computer Science and Systems Engineering. Lecturer at the Department of Industrial and Systems Engineering, Faculty of Science and Engineering, Hosei University, Japan.
My research concerns statistical methods for drawing correct conclusions from small samples: Bayesian estimation (noninformative priors and optimal estimators), reliability engineering (accelerated life tests and degradation models), and item response theory (adaptive testing).
Research Topics
- Bayesian estimation with noninformative priors
- Deriving optimal Bayesian estimators under reference and Jeffreys priors for various distributions.
- Reliability, degradation and accelerated life tests
- Predicting product lifetime with statistical models from accelerated tests and degradation data.
- Item response theory and adaptive testing
- Test designs and estimation methods that measure ability accurately with few items.
- Statistical modeling of real data
- Analyzing data on infectious diseases, parts demand, purchases and sports with statistical models.
Recent Papers (FY2024–FY2026)
- Ko Abe, Takenori Sakumura (2026) “Empirical Joint Estimation for Incubation Period and Infection under the Doubly Interval-Censored with Right Truncated Observation,” Journal of Computational and Graphical Statistics. [DOI]
- 今泉智子，作村建紀，岡田謙介 (2026) 「密度比に基づいた Haebara 法の拡張等化手法」 大学入試研究ジャーナル, 36, 114–121.
- Takenori Sakumura, Takemi Yanagimoto (2025) “Optimal Bayesian estimator under the reference prior in the von Mises distribution,” Communications in Statistics-Simulation and Computation, 1–13, (Online First). [DOI]
- Takenori Sakumura, Takemi Yanagimoto (2025) “Bayesian Estimation under the Conjugate-Gamma Prior for the Weibull Distribution,” Proceedings of the Institute of Statistical Mathematics, 73, 253–266, (in Japanese).
- Takenori Sakumura, Takemi Yanagimoto (2024) “Very Small Bias Observed in Bayesian Estimators even for Small Sample Sizes,” Communications in Statistics - Simulation and Computation, 53(11), 5594–5604. [DOI]
Corporate Training
I offer in-house training in statistics and reliability engineering: Weibull analysis and accelerated degradation tests, Bayesian statistics, and item response theory, using data from quality assurance, testing and HR education (in Japanese).
Contact
- E-mail: sakumura (at) hosei.ac.jp
- Address: Room W6004, 6th floor, West Building, Koganei Campus, Hosei University, 3-7-2 Kajino-cho, Koganei, Tokyo 184-8584, Japan (map)
Laboratory
Undergraduate and graduate students study statistical modeling and data analysis in the lab.