Journal of the Japanese Society of Computational Statistics
Online ISSN : 1881-1337
Print ISSN : 0915-2350
ISSN-L : 0915-2350
Volume 14, Issue 1
Displaying 1-7 of 7 articles from this issue
  • Yuichi Takeda
    2001 Volume 14 Issue 1 Pages 1-10
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    We consider the problem of testing the equality of intermediate characteristic roots in two populations. The permutation test is investigated for testing the hypothesis. The exact distribution of the ratio of the largest characteristic roots across populations is derived under the assumption of multivariate normality. A Monte Carlo experiment is conducted to examine the performance of the permutation test under the assumptions that two population distributions are characterized by multivariate normal and contaminated multivariate normal distributions.
    Download PDF (826K)
  • Masahiro Kuroda, Zhi Geng, Naoto Niki
    2001 Volume 14 Issue 1 Pages 11-29
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    In this paper, we discuss the Bayesian sequential learning on probabilities from incomplete data in decomposable graphical models. We give exact formulas of the posterior distribution, and the posterior mean and the posterior second moment based on a hyper Dirichlet prior distribution and an incomplete observation. The posterior distribution is usually a mixture hyper Dirichlet distribution when there exist incomplete data. In order to approximate the mixture posterior, we choose a single hyper Dirichlet distribution which has the same mean and the same average variance sum as those of the exact posterior.
    Download PDF (1560K)
  • Ken Nittono, Toshinari Kamakura
    2001 Volume 14 Issue 1 Pages 31-47
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    A modified method for Bayesian image restoration using varying neighborhood structure is proposed. The method reduces computational burden for yielding a restored image due to the dynamical change of structural forms of neighborhood, which should be iteratively and adaptively composed through the process of the restoration calculation. Although, in practice, the results of restoration generally depend on given data, our simulation results show that the method is effective for some given gray-scale images with moderate additive Gaussian noise.
    Download PDF (5784K)
  • Taka-aki Shiraishi
    2001 Volume 14 Issue 1 Pages 49-57
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    Statistical estimation procedures are proposed based on studentized robust statistics for location parameters in two-way layouts with interaction. Large sample properties of these procedures as the cell size tends to infinity are investigated. Although Fisher's consistency is assumed in the theory of M-estimators, it is not needed in this paper. It is found that the asymptotic relative efficiencies (ARE's) of the proposed procedures relative to classical procedures agree with the classical ARE-results of Huber's one sample M-estimator relative to the sample mean. By simulation studies, it can be seen that the proposed estimators are more efficient than least squares estimators except for the case where the underlying distribution is normal.
    Download PDF (839K)
  • Takeshi Fujiwara, Junji Nakano, Yoshikazu Yamamoto, Ikunori Kobayashi
    2001 Volume 14 Issue 1 Pages 59-67
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    As computing technologies have been developed rapidly and continuously, statistical languages should be frequently improved by adopting them for realizing comfortable statistical environments. Although that work was a difficult task before, we can implement such languages easily by modern software techniques such as the Java language, which unifies recent computer technologies widely and neatly.
    In this paper, we describe the idea of a new Jasp (JAva based Statistical Processor) language, which is a function-based and object-oriented language, is suitable for interactive operation, and has flexibilities and diverse extensibilities. Considering required characteristics for a statistical language, we decide to implement the Jasp language by Java technologies through the Pnuts language, which is a script language written in Java, and a preprocessing approach. This implementation brings several advantages such as reduction of development costs, effective uses of existing resources, and adequate reliabilities.
    Download PDF (2861K)
  • 2001 Volume 14 Issue 1 Pages 69-73
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    Download PDF (338K)
  • 2001 Volume 14 Issue 1 Pages 74-78
    Published: 2001
    Released on J-STAGE: December 09, 2009
    JOURNAL FREE ACCESS
    Download PDF (389K)
feedback
Top