文献一覧: 03895602 (ISSN) - Ceek.jp Altmetrics

1 0 0 0 OA INFORMATION INEQUALITIES FOR THE MINIMAX RISK

著者: Sato Michikazu Akahira Masafumi
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.25, no.2, pp.151-158, 1995

This paper presents lower bounds for the minimax risk under quadraticloss, derived from information inequalities for the Bayes risk obtained byBorovkov and Sakhanienko, Brown and Gajek. In addition, admissibilityof a minimax estimator is discussed, and we provide examples which illustratethat they are good bounds.

2013-03-01 08:31:54
1 + 0 Twitter

http://hdl.handle.net/2241/118444

1 0 0 0 OA A NEW HIGHER ORDER APPROXIMATION TO A PERCENTAGE POINT OF THE DISTRIBUTION OF THE SAMPLE CORRELATION COEFFICIENT

著者: Akahira Masafumi Torigoe Norio
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.28, no.1, pp.45-57, 1998

A new higher order approximation formula for a percentage point of thedistribution of the sample correlation coefficient is given up to the order O( n-1),using the Cornish-Fisher expansion for the statistic based on a linear combinationof a normal random variable and chi-random variables. The numerical comparisonof the formula with others shows that it dominates the others and gives almostprecise values in various cases even for the size n= 10 of sample.

2013-03-01 08:31:48
1 + 0 Twitter

http://hdl.handle.net/2241/118439

1 0 0 0 OA THE COMPARISON OF ESTIMATORS OF RATIO FOR A REGRESSION MODEL

著者: Kawai Shinichi Akahira Masafumi
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.24, no.2, pp.141-150, 1994

In some regression model, the mean square' errors of a ratio estimator, a groupedjackknife estimator, and an estimator based on the least square estimators (LSEs) areobtained and compared up to the order O(n-3), where n is the size of the sample. Thebias-adjusted ratio estimator and the jackknife estimator are also compared up to theorder O(n-3). Then it is concluded that the estimator based on the LSEs is an asymptoticallybetter estimator of ratio up to the order o (n-3).Some examples are given.

2013-03-01 08:31:26
1 + 0 Twitter

http://hdl.handle.net/2241/118440

1 0 0 0 OA THE OPTIMALITY OF THE GROUPED JACKKNIFE ESTIMATOR OF RATIO IN SOME REGRESSION MODEL

著者: Akahira Masafumi Kawai Shinichi
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.20, no.2, pp.149-157, 1990

In some regression model, the minimum (asymptotic) variance estimator of a ratiois discussed for some class of linear combinations of ratio estimators, and the jackknifeprocedure is considered. It is seen that the grouped jackknife estimator is optimal inthe sense that it has asymptotically the minimum variance in the class. Higher orderbias reduction of the estimators is discussed, and some examples are given.

2013-03-01 08:30:59
1 + 0 Twitter

http://hdl.handle.net/2241/118427

1 0 0 0 OA BEHAVIOUR OF JACKKNIFE ESTIMATORS IN TERMS OF ASYMPTOTIC DEFICIENCY UNDER TRUE AND ASSUMED MODELS

著者: Akahira Masafumi
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.19, no.2, pp.179-196, 1989

The problem on jackknifing estimators is investigated in the presence of nuisanceparameters from the viewpoint of higher order asymptotics. It is shown that theasymptotic deficiency of the jackknife estimator relative to the bias-adjusted maximumlikelihood estimator (MLE) is equal to zero under true and assumed m.odcls. Moreover,the asymptotic deficiency of the MLE or the jackknife estimator under the assumedmodel relative to that under the true model is given.

2013-03-01 08:30:53
1 + 0 Twitter

http://hdl.handle.net/2241/118426

1 0 0 0 OA ON THE NEW APPROXIMATION TO NON-CENTRAL T-DISTRIBUTIONS

著者: Akahira Masafumi Sato Michikazu Torigoe Norio
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.25, no.1, pp.1-18, 1995

Recently a new approximation to a percentage point of non-centralt-distributions was proposed by Akahira [1]. In this paper the approximationformula is presented and the existence and uniqueness of a solution of theequation on the formula is proved. Numerical results are also given.

2013-03-01 08:30:37
1 + 0 Twitter

http://hdl.handle.net/2241/118436

1 0 0 0 OA THE STRUCTURE OF THE ASSUMED MODEL THROUGH THE DISCRETIZED LIKELIHOOD ESTIMATOR

著者: Akahira Masafumi
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.23, no.1, pp.19-31, 1993

In the presence of a nuisance parameter the asymptotic deficiency of the discretizedlikelihood estimator (DLE) relative to the bias-adjusted maximum likelihood estimatoris obtained under the assumed model. It consists of two parts. One is the lossof information associated with the DLE of the parameter to be estimated. Another,is that due to the "incorrectness" of the assumed model. Some examples on the normaland Weibull type distributions are given.

2013-03-01 08:30:26
1 + 0 Twitter

http://hdl.handle.net/2241/118432

1 0 0 0 OA A HIGHER ORDER LARGE-DEVIATION APPROXIMATION FOR THE DISCRETE DISTRIBUTIONS

著者: Akahira Masafumi Takahashi Kunihiko
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.31, no.2, pp.257-267, 2001-12

For a sum of independent discrete random variables, its higher order large-deviation approximation is discussed. An approximation to the tail probability of the distribution of the sum is provided, and its numerical comparison with other approximations is done in the binomial case. Consequently, the approximation formula is seen to be more accurate.

2013-02-28 08:30:23
1 + 0 Twitter

http://hdl.handle.net/2241/118421

1 0 0 0 OA WEIGHTED LOSS FUNCTIONS FOR SET ESTIMATION AND TESTING HYPOTHESES

著者: Maihara Hirosuke Akahira Masafumi
出版者: 日本統計学会
雑誌: Journal of the Japan Statistical Society (ISSN:03895602)
巻号頁・発行日: vol.34, no.2, pp.189-206, 2004-12

From the decision-theoretic viewpoint, using a weighted loss we compare the risks of testing procedures in the location and scale parameter cases. We also get numerically the minimax solution of Bayes testing procedures w. r. t. a parameter of the prior distribution, under the weighted loss.