Kodo Keiryogaku (The Japanese Journal of Behaviormetrics)
Online ISSN : 1880-4705
Print ISSN : 0385-5481
ISSN-L : 0385-5481
Volume 36 , Issue 2
Showing 1-5 articles out of 5 articles from the selected issue
Article
  • Hiroko NAKANISHI
    2009 Volume 36 Issue 2 Pages 67-80
    Published: 2009
    Released: June 29, 2010
    JOURNALS FREE ACCESS
    Compositional data consist of vectors of positive percentage values summing up to a unit. In this paper, we propose two variable selection rules for compositional data and two criteria for a recommended solution to select a subset of variables. The proposed rules are based on minor determinants of a variance-covariance matrix or a correlation matrix. They are the same as McCabe's (1984) idea for variable selection rules in principal component analysis. We apply the rules and the criteria to three real datasets and consider their performance. As a conclusion, the performance of the rule based on a minor determinant of a variance-covariance matrix is better than the other one. The rule is useful to apply, but further consideration would be necessary for applying the criteria to any datasets.
    Download PDF (305K)
Notes
  • Masao UEDA
    2009 Volume 36 Issue 2 Pages 81-88
    Published: 2009
    Released: June 29, 2010
    JOURNALS FREE ACCESS
    The understanding of consumer product and brand image is vital to the implementation of marketing. The understanding of consumer image through associative networks in particular allows for a deep understanding of the relationships between associations and provides effective business hints. In this study a new research method for understanding consumer images through associative networks was proposed and an empirical study was held to determine the effectiveness of this proposed method. Analyzing the survey results showed that the associations obtained by having participants create associative networks are sufficient for use in marketing purposes, as intended.
    Download PDF (358K)
  • Ming-Zhe JIN
    2009 Volume 36 Issue 2 Pages 89-103
    Published: 2009
    Released: June 29, 2010
    JOURNALS FREE ACCESS
    In this research, as a basis of studies regarding when certain works were written, an estimation was attempted using the works of Ryunosuke Akutagawa. In the experiment, two types of data sets were created from the text with part-of-speech tagging, and a comparative analysis was performed using three methods: Linear Regression, Support Vector Regression, and Random Forest Regression. As a result, when the works were written was estimated with rather high accuracy. The average of absolute value of estimation error and standard deviation was approximately 1.4 years. The order of high accuracy of estimation was Random Forest Regression, Support Vector Regression, and Linear Regression.
    Download PDF (1174K)
  • Shinji DEGUCHI, Akinori OKADA, Atsuho NAKAYAMA
    2009 Volume 36 Issue 2 Pages 105-114
    Published: 2009
    Released: June 29, 2010
    JOURNALS FREE ACCESS
    When we use multidimensional scaling (MDS) to analyze a dataset in which the magnitude of the values considerably differ (vary) among objects, for example, customer purchase history data, it often occurs that the only points of objects with large values scatter well in the configuration, but the others with small values do not scatter well. It is said that joint uses of MDS and cluster analysis is often desirable. In this study, we showed that the joint use of MDS and cluster analysis also worked well in the above case, through the analysis of two-mode three-way proximity data such as several variance-covariance matrices of objects obtained from each condition. We also used an external analysis with MDS to show the information about each of the two modes on the same configuration simultaneously, in order to easily observe the global structure of data through the configuration.
    Download PDF (375K)
Book Review
feedback
Top