Chooichiro Asano

Bayesian estimation methods for categorical data with misclassifications

This article considers Bayesian estimation methods for categorical data with misclassifications. To adjust for misclassification, double sampling schemes are utilized. Observations are represented in a contingency table categorized by error-free categorical variables and error-prone categorical variables. Posterior means of probabilities in cells are considered as estimates. In some cases, the posterior means can be calculated exactly. However,in some cases, the exact calculation may be too difficult to perform, but we can easily use the expectation-maximiza-tion(EM) algorithm to obtain approximate posterior means.

Runs test for a circular distribution and a table of probabilities

SummaryA method is suggested for testing whether two samples observed on a circle are drawn from the same distribution. The proposed test is a modification of the well-known Wald-Wolfowitz runs test for a distribution on a straight line. The primary advantage of the proposed test is that it minimizes the number of assumptions on the theoretical distribution.

Analysis of ordered categorical data from repeated measurements assuming a quantitative latent variable

SummaryThe purpose of the present paper is to propose an analytical method for ordered categorical responses obtained from a repeated measurement/longitudinal experiment. The ordered categorical scale is assumed to be a manifestation of a latent quantitative variable. A linear model is assumed for location parameters of the underlying distributions. Weighted least square method is applied to parameter estimation and subsequent analysis. Two data sets are analyzed to show several aspects of analysis by the proposed model and to discuss comparative characteristics of analysis compared with earlier analysis. A mention is made for a computer software program for the proposed model.

Latent scale linear models for multivariate ordinal responses and analysis by the method of weighted least squares

SummaryA multivariate latent scale linear model is defined for multivariate ordered categorical responses and inference procedures based on the weighted least squares method are developed. Several applications of the model are suggested and illustrated through an analysis of real data. Asymptotic properties of the weighted least squares method are examined and some consequences of misspecification of the model are also discussed.

Web-based statistical system by using the DLL

Abstract The requirements for statistical software vary with the users’ goals and environmental experience. Although many developers have attempted to create their software for such various demands, it is virtually difficult to develop and offer such a system with a wide variety of users. The offered system could give a solution to this problem. The basic idea is to separate the interface from statistical engine and to apply the web browser as the interface and Dynamic Link Library (DLL) as statistical library. The application of a web browser has great advantages as an interface, e.g. ability of addition of documents, graphical objects, hyperlinks to other relative sites on the web, and so on. These capabilities make it easy to develop an interface to fit users’ wide needs by using the system. Combining a web-based interface page with an appropriate site by hyperlinks, the users can access the site quickly and analyze data while reading related documents.

A New Statistical Tool for Finding Causal Conditions for Small Data Sets and Its Software

We propose a method that finds target groups such that the response probability of a dichotomous dependent variable is larger than a pre-specified target value. This is enabled by a selection of independent variables with high order interactions. The method is effective to extract interesting patterns from datasets. We developed software for this method. The software incorporates various types of tools like data management, model building, visualization, and so on.