• Communications for Statistical Applications and Methods
• /
• v.3 no.1
• /
• pp.11-19
• /
• 1996

Canonical correlation analysis is a multivariate technique for identifying and quantifying the statistical relationship between two sets of variables. Like most multivariate techniques, the main objective of canonical correlation analysis is to reduce the dimensionality of the dataset. It would be particularly useful if high dimensional data can be represented in a low dimensional space. In this study, we will construct statistical graphs for paired sets of multivariate data. Specifically, plots of the observations as well as the variables are proposed. We discuss the geometric interpretation and goodness-of-fit of the proposed plots. We also provide a numerical example.

• Communications for Statistical Applications and Methods
• /
• v.26 no.3
• /
• pp.305-313
• /
• 2019

Generalized canonical correlation analysis (GCCA) extends the canonical correlation analysis (CCA) to the case of more than two sets of variables and there have been many studies on how two-set canonical solutions can be generalized. In this paper, we derive certain stationary equations which can lead the higher-order solutions of several GCCA methods and suggest a type of iterative procedure to obtain the canonical coefficients. In addition, with some numerical examples we present the methods for graphical display, which are useful to interpret the GCCA results obtained.

• Journal of the Korean Statistical Society
• /
• v.31 no.2
• /
• pp.163-175
• /
• 2002

An extended version of the minimax eccentricity factor estimation for multiple set case is proposed. In addition, two more simple methods for multiple set factor analysis exploiting the concept of generalized canonical correlation analysis is suggested. Finally, a certain connection between the generalized canonical correlation analysis and the multiple set factor analysis is derived which helps us clarify the relationship.

• Women's Health Nursing
• /
• v.14 no.3
• /
• pp.213-221
• /
• 2008

Purpose: This study aimed to examine the accuracy and adequacy of research papers reporting statistical testings for correlation and regression. Method: Original research articles utilized correlation and regression analysis were reviewed from the Korean Journal of Women Health Nursing published from the year 2004 to 2006. Thirty-six papers were evaluated in accordance with formatted criteria in respect to an inclusiveness of research title, accuracy of statistical methods and presentation styles, and errors in reporting statistical outcomes. Result: Thirty articles (83.3%) utilized Pearson's correlational analysis, and ten articles did regression analysis. Lack of accurate understanding and interpretation of the statistical method was a main fault. Basic assumptions and diagnostic testings for each statistical method were not performed or described in most of the studies. Some points like consistency of research questions with statistical methods and criteria for sample size were still left out in part. Details of the presentation in the reporting of outcomes were not complied with the guidelines, which need careful concerns of the writers. Errors in English of result tables were found in more than one third of the tables. Conclusion: The outcome would be reflected in the submission guidelines for future writers. To reach the level comparable with internationally recognized nursing journals, concrete knowledge to apply statistical methods should be ensured in the processes of submission, reviews, and editing.

• Communications for Statistical Applications and Methods
• /
• v.19 no.3
• /
• pp.471-478
• /
• 2012

In this paper, we present a permutation test to select the number of pairs of canonical variates in canonical correlation analysis. The existing chi-squared test is known to be limited to normality in use. We compare the existing test with the proposed permutation test and study their asymptotic behaviors through numerical studies. In addition, we connect canonical correlation analysis to regression and we we show that certain inferences in regression can be done through canonical correlation analysis. A regression analysis of real data through canonical correlation analysis is illustrated.

• Journal of Korean Society for Quality Management
• /
• v.18 no.1
• /
• pp.84-95
• /
• 1990

The objective of this study is to understand the effect of measurement-related investment. That is, this study aims at verifying the correlation between the measurement-related investment and inferiority rate of products by statistical analysis. The samples of this study are 376 industrial companies in Korea, and the research data was analysed on inferiority state of industrial companies with a measurement standard laboratory. The analysis was made by the elementary statistics, the correlation analysis and the regression analysis. The results are summarized as follows : First, the inferioriy rate of the industrial companies with a measurement standard laboratory was relatively lower than that of the other companies without the laboratory by statistical significance. Second, the increment on measurement-related investment had a negative correlation with the increment of inferiority rate, and the increase of measurement-related investment showed decrease of the inferiority rate by regression analysis.

• Women's Health Nursing
• /
• v.17 no.2
• /
• pp.187-194
• /
• 2011

Purpose: This study investigated the statistical methods and the results had reported correlation/regression analysis in the studies of Korean Journal of Women Health Nursing (KJWHN). Methods: We reviewed 45 studies using correlation/regression analysis for the suitability of the statistical methods and the research purposes, the criteria for analysis of figures, tables and charts had published in the KJWHN from vol 13 (1) in 2007 to vol 15 (4) in 2009. Results: Forty three studies were fitted to their statistical methodology and their research purposes. Eleven studies considered the minimum sample size. Fourteen regression studies used multiple regression and 12 studies used forward method for variable entry. Only one study among the 17 regression studies accomplished scatter plots and residuals examination. Sixteen studies in correlation studies and six studies in regression studies showed some errors in either the title, variables, category of figures, tables and charts. In the regression study, all reported $R^2$ and ${\beta}$ values except one. Conclusion: It was found that there were still statistical errors or articulation errors in the statistical analysis. All reviewers need to be reviewed more closely for detecting errors not only during reviewing process of the manuscript but also periodic publication for the quality of this academic journal.

• Journal of the Korean Statistical Society
• /
• v.25 no.4
• /
• pp.589-601
• /
• 1996

Geometric approach to extend the classical two-set theory of canonical correlation analysis to three or more sets is considered. It provides statistical graphs to represent the data in a low dimensional space. Procedures are developed for computing the canonical variables and the corresponding properties are investigated. The solution is equivalent to that of the usual problem in the case of two sets. Goodness-of-fit of the proposed plots is studied and a numerical example is included.

• Communications for Statistical Applications and Methods
• /
• v.10 no.3
• /
• pp.695-705
• /
• 2003

In this article, we propose a regression and correlation analysis via dynamic graphs and implement them in Java Web Start. For the polynomial relations between dependent and independent variables, dynamic graphics are implemented for both polynomial regression and spline estimates for an instant model selection. The results include basic statistics. They are available both as a web-based service and an application.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.387-397
• /
• 2003

Generally, most of researches that need a variable-clustering process use an exploratory factor analysis technique or a divisive hierarchical variable-clustering method based on a correlation matrix. And some researchers apply a object-clustering method to a distance matrix transformed from a correlation matrix, though this approach is known to be improper. On this paper an agglomerative hierarchical variable-clustering method based on a correlation matrix itself is suggested. It is derived from a geometric concept by using variate-spaces and a characterizing variate.