発表

1A-039

期待平方誤差を用いた分散共分散の予測推定

[責任発表者] 小笠原 春彦:1
1:小樽商科大学

Predictive estimators (Ogasawara, 2017ab, 2018b) for variances and covariances can be derived using associated mean square errors. In this paper, predictive estimation of unconstrained variances and covariances is considered.

References
Ogasawara, H. (2015). Bias adjustment minimizing the asymptotic mean square error. Communications in Statistics – Theory and Methods, 44, 3503-3522.
Ogasawara, H. (2017a). A family of the adjusted estimators maximizing the asymptotic predictive expected log-likelihood. Behaviormetrika, 44, 57-95.
Ogasawara, H. (2017b). Predictive estimation of a covariance matrix and its structural parameters. Journal of the Japanese Society of Computational Statistics, 30, 45-63.
Ogasawara, H. (2018a). Supplement to the paper “Predictive estimation of a covariance matrix and its structural parameters”. Economic Review (Otaru University of Commerce), 69 (1), 1-17. http://www.res.otaru-uc.ac.jp/~emt-hogasa/.
Ogasawara, H. (2018b). An asymptotic equivalence of the cross-data and predictive estimators. Communications in Statistics -Theory and Methods (on-line published). https://doi.org/10.1080/03610926.2018.1549258.

キーワード
標本共分散の母集団共分散/非正規分布/4次のキュミュラント


詳細検索