Welcome to 

Takuma Yoshida's Home Page

Name: Takuma Yoshida
Institution: Graduate School of Science and Engineering, Kagoshima University, Japan. 
Address: 1-21-35, Korimoto, Kagoshima, Japan 
Position: Associate Professor
E-mail: yoshida[at]sci.kagoshima-u.ac.jp
Research Interest:  Extreme value theory, Nonparametric regression, Small area estimation, Sparse modeling, 
Working paper:・Momoki.K and Yoshida.T. Unit-level mixed effects models for conditional extremes . Submitted. arXiv ・Yoshida.T.   Asymptotic theory for extreme value generalized additive models. Submitted. arXiv ・Yoshida.T.  Single-index models for extreme value index regression. submitted. Submitted. arXiv・Yoshida.T.  Variable screening in extremal quantile regression. preprint.  
Accepted paper (selected): ・Momoki.K and Yoshida.T. (2024). Hypothesis testing for varying coefficient models in tail index regression . Statistical Papers. https://doi.org/10.1007/s00362-024-01538-0・Iwaanakuchi, T., Yoshida. T., Fukuda, Y., and Uto, Y. (2023).  Impact of cognitive decline on medical outcomes and nursing workload: A retrospective cohort study. Plos One. 18. e0293755. ・Homma.G. and Yoshida.T. (2023). Sample size calculation in clinical trials with two co-primary endpoints including overdispersed count and continuous outcomes.  Pharmaceutical Statistics,23. 46-59.・Kawasaki. H., Fudamoto.K., Yamamoto,M., Iwaanakuchi,T., Yoshida. T., Hashiguchi,T. and Uto.Y. (2023). Verification of the relationship between the sequential organ failure assessment score and the length of intensive care unit and hospital stay in terms of medical resources input . Medicine. 102. e34632.・Yoshida. T.  (2021) . Quantile function regression and variable selection for sparse models. Canadian Journal of Statistics. 49. 1196 - 122.・Yoshida.T. (2021).  Simultaneous confidence bands for extremal quantile regression with splines.  Extremes. 23.   117 - 149.・Yoshida.T. (2020). Two stage smoothing in additive models with missing covariates.  Statistical Papers. 60   1803 - 1826.・Yoshida.T. (2020).  Extreme value inference for quantile regression with varying coefficients .  Communications in Statistics - Theory and Methods. 50 . 685 - 710 ・Sasaki.M., Uto.Y., Yoshida.T., Iwaanakuchi,T., Muranaga,F., Saigo,Y., Kumamoto,I.  (2019).  Secondary use of hospital information system data for safe bedside radiography in terms of patient factors. .  Health information management : Journal of the Health Information Management Association of Australia. 48. 24 - 32.・Yoshida. T. and Naito.K. (2019).  Regression with stagewise minimization on risk function.  Computational Statistics and Data Analysis. 134   123 - 143.・Yoshida.T. (2018).  Semiparametric method for model structure discovery in additive regression models.  Econometrics and Statistics Part B: Statistics. 5. 124 - 136.・Yoshida. T. and Naito. K. (2014).  Asymptotics for penalized splines in generalized additive models. Journal of Nonparametric Statistics. 26. 269 - 289. ・Yoshida.T. and Naito. K. (2012). Asymptotics for penalized additive B-spline regression. Journal of the Japan Statistical Society. 42. 81-107.・Yoshida,T., Kanba,M., and Naito,K. (2010). A computationally efficient model selection in the generalized linear mixed model. Computational Statistics. 25. 463-484.

International Conference (selected): 

・Yoshida.T.  Asymptotic theory for extreme value generalized additive models.  The 10th International Congress on Industrial and Applied Mathematics. Tokyo, Japan. August 2023. 

・Momoki.K. and Yoshida.T.  Mixed effects models for large sized clustered extremes. The 6th International Conference on Econometrics and Statistics (EcoSta), Tokyo, Japan, August 2023.

・Yoshida,T.  Asymptotic theory for extreme value generalized additive models.  The 13th International Conference on Extreme Value Analysis. Milan, Italy. June 2023. 

・Yoshida,T. Simultaneous confidence bands for extremal quantile regression with splines . The 11th International Conference on Extreme Value Analysis. Zagreb, Croatia. June 2019.   

・Yoshida,T. Extreme Inference for Nonparametric Quantile Regression with Heavy Tailed Data .  Joint Statistical Meeting 2018 . Vancouver, Canada. August 2018.

・Yoshida,T. Semiparametric estimation for model structure discovery . CMStatistics 2015. London, England. December 2015.

Education

・Ph.D (Science). Graduate School of Science and Engineering, Shimane University, Japan. March, 2013. 

・M.A (Science). Graduate School of Science and Engineering, Shimane University, Japan. March 2010.

・B.A. (Science). Department of Science and Engineering, Shimane University, Japan. March 2008.

Employment

・October, 2018-present. Associate Professor, Graduate School of Science and Engineering, Kagoshima University, Japan.

・April, 2013-September, 2018. Assistant Professor, Graduate School of Science and Engineering, Kagoshima University, Japan.

Grants:・JSPS  Grant-in-Aid for Scientific Research C 2022-2025・JSPS Grant-in-Aid for Early-Career Scientists  2018-2022・JSPS Grant-in-Aid for Young Scientists  2014-2018