Masahiro Mizuta

Hokkaido University | Hokudai · Advanced Data Science Laboratory, Information Initiative Center

Functional data analysis has been proposed by Ramsay and Silverman, and since then, many theories and methods have been developed. We have already proposed discrete functional analysis (Mizuta 2006). However, in this paper, we refine it and provide a theoretical explanation of uncorrelated discrete differences. An effective approach in functional d...

Background: Concerns have been raised about the adverse health impacts of mobile device usage. The objective of this cross-sectional study was to examine the association between a child’s age at the first use of a mobile device and the duration of use as well as associated behavioral problems among school-aged children. Methods: This study focused...

In this paper, we propose a method for extracting nonlinear structure from multi-dimensional data. In dimension reduction such as principal component analysis (PCA) and projection pursuit (PP) (Friedman in J Am Stat Assoc 82(397):249–266, 1987), we search for projection directions which maximize an index, variance (PCA) or projection indices (PP)....

The opportunities for exposure to radiofrequency electromagnetic fields (RF-EMF) among children are increasing. Children's exposure to RF-EMF in Japan was recorded using a personal exposure meter (ExpoM-RF), and factors associated with the exposure examined. A total of 101 children, aged 10–15 years old, participated in the prospective birth cohort...

Serum fatty acids (FAs) exist in the four lipid fractions of triglycerides (TGs), phospholipids (PLs), cholesteryl esters (CEs) and free fatty acids (FFAs). Total fatty acids (TFAs) indicate the sum of FAs in them. In this study, four statistical analysis methods, which are independent component analysis (ICA), factor analysis, common principal com...

With increasing use of mobile phones, exposure to radiofrequency electromagnetic field (RF-EMF) in the high-frequency band associated with mobile phones has become a public concern, with potentially adverse effects on cognitive function in children and adolescents. However, findings regarding the relation of RF-EMF and cognitive function in childre...

China has become the world’s second largest economy after recent rapid economic growth. On the other hand, the urbanization of China has aggravating the problems of air pollution, traffic congestion, and the dual structure of urban and rural areas. Therefore, the Chinese government has implementing a policy to revise the efficiency-oriented city pl...

We often meet the case in data analysis that the explanatory variables can be occasionally divided into two groups. One group comprises the variables that researchers consider controllable, and the other group comprises those they do not. We call them controllable and uncontrollable variables, respectively. In the study, we deal with binary respons...

In this research, we study characteristics of concentration variation with functional data analysis. Functional data analysis, which combines traditional data analysis with the characteristics of functions, is suitable to analyze changing trend of observed data with the utility of derivatives. We applied functional clustering analysis to milk datas...

A novel functional clustering method for domain-dependent functional data is proposed. The functional data are defined on sequential domains and they might have particular clusters on each domain. Our approach, Moving Functional k-means Clustering, is to classify the peculiar domains based on each sequential set of clusters. A typical functional cl...

Cholesteryl ester (CE) is an ester of cholesterol and fatty acid (FA). Plasma CE reflects complicated metabolisms of cholesterol, phospholipids, lipoproteins, and dietary FAs. An informatics approach could be useful for analysis of the CE species. In this study, two basic dimension reduction methods, principal component analysis (PCA) and factor an...

Purpose We provide details of the dual-focal 3D tracking and pixel intensity calibration (DuFT) system used to record gene expression in targeted areas on the skin of freely moving mice. Methods The accuracy of the 3D position calculation was determined by placing a scintillator on the calibration disk at various known locations within the recordi...

25-Hydroxyvitamin D3 (25(OH)D3) as the metabolite of vitamin D, is connected with various of diseases, and important to people with limited sunshine. Thus, the investigation of serum 25-hydroxyvitamin D and its variation in these people is necessary. In this study, a simple, precise, and accurate method for serum 25(OH)D3 determination by LC/MS/MS...

In this paper, we compare two radiation effect models: the average surviving fraction (ASF) model and the integral biologically effective dose (IBED) model for deriving the optimal irradiation scheme and show the superiority of ASF. Minimizing the effect on an organ at risk (OAR) is important in radiotherapy. The biologically effective dose (BED) m...

We studied change in the plasma total, esterified and non-esterified capric acid (FA10:0) and its effect on longer fatty acid concentrations during the short-term oral administration of synthetic tricaprin in dogs. We administered 150 and 1500 mg tricaprin/kg body weight per day orally to dogs for 7 consecutive days. Blood samples were collected at...

Motivation: DNA methylation is an important epigenetic modification related to a variety of diseases including cancers. We focus on the methylation data from Illumina's Infinium HumanMethylation450 BeadChip. One of the key issues of methylation analysis is to detect the differential methylation sites between case and control groups. Previous appro...

Purpose: Radiotherapy of solid tumors has been performed with various fractionation regimens such as multi- and hypofractionations. However, the ability to optimize the fractionation regimen considering the physical dose distribution remains insufficient. This study aims to optimize the fractionation regimen, in which the authors propose a graphic...

The future of radiotherapy will be composed of absorbed dose in the four-dimensional coordinates of time and space with an emphasis on optimizing spatial dose distribution and temporal dose distribution. We refer to this concept as “dose composition radiotherapy (DCRT)” and anticipate that future generations of radiation therapy will incorporate th...

In this paper, we propose hierarchical cluster analysis and multidimensional scaling for joint distribution valued data. Information technology is increasing the necessity of statistical methods for large and complex data. Symbolic Data Analysis (SDA) is an attractive framework for the data. In SDA, target objects are typically represented by aggre...

There are increasing requirements for analysing very large and complex datasets derived from recent super-high cost performance computer devices and its application software. We need to aggregate and then analyze those datasets. Symbolic Data Analysis (SDA) was proposed by E. Diday in 1980s (Billard L, Diday E (2007) Symboic data analysis. Wiley, C...

Purpose The authors propose a graphical representation of the relation between the effect on the tumor and the damage effect on an organ at risk (OAR) against the irradiation dose, as an aid for choosing an appropriate fractionation regimen. Methods The graphical relation is depicted by the radiation effect on the tumorE1 versus that on an OAR E0....

Hypofractionated irradiation is often used in precise radiotherapy instead of conventional multifractionated irradiation. We propose a novel mathematical method for selecting a hypofractionated or multifractionated irradiation regimen based on physical dose distribution adding to biologic consideration. The linear-quadratic model was used for the r...

One characteristic of computational statistics is the processing of enormous amounts of data. It is now possible to analyze large amounts of highdimensional data through the use of high-performance contemporary computers. In general, however, several problems occur when the number of dimensions becomes high. The first problem is an explosion in exe...

We deal with methods for analyzing complex structured data, especially, distribution valued data. Nowadays, there are many requests to analyze various types of data including spatial data, time series data, functional data and symbolic data. The idea of symbolic data analysis proposed by Diday covers a large range of data structures. We focus on di...

In this paper, we discuss sensitivity analysis in linear subspace method, especially on multiple-case diagnostics. Linear subspace method by Watanabe (1973) is a useful discriminant method in the field of pattern recognition. We have proposed its sensitivity analyses, with single-case diagnostics and multiple-case diagnostics with PCA. We propose a...

In this paper, we deal with the relation between functional clustering and functional principal points. The k principal points are defined as the set of k points which minimizes the sum of expected squared distances from every points in the distribution to the nearest point of the set, and are mathematically equivalent to centers of gravity by k-me...

In this paper, we deal with functional data analysis including functional clustering and an application of functional data analysis. Functional data analysis is proposed by Ramsay et al. In functional data analysis, observed objects are represented by functions. We give an overview of functional data analysis and describe an actual analysis of Mu...

In this paper, we try to extend the framework of Functional Data Analysis (FDA). FDA is an exciting theme that continues development in data analysis. We can sometimes find out valuable information through FDA. Most methods on FDA assume that the functions that represent data are differentiable. But we discuss the use of discrete functions for FDA....

We deal with graphical representations of results of functional clustering and functional multidimensional scaling (MDS). Ramsay and Silverman(1997, 2005) proposed functional data analysis. Functional data analysis enlarges the range of statistical data analysis. But, it is not easy to represent results of functional data analysis techniques. We fo...

In this paper, we extend relative projection pursuit proposed by Mizuta[Miz02] for functional data and offer a new method on functional data analysis[Ram82, RS97], which called functional relative projection pursuit. The aim is to find 'in-teresting' projections of functional data, for example, to find structures that are separated by groups. The p...

We consider the problem of finding out the features of the data, when a data set is divided into the groups in advance. Flury (1984) proposed Common Principal Components Analysis (CPC) which assumes that the covariance matrices have a common basis structure to solve the problem. In this article, we extend CPC model and to propose a new common princ...

Radial basis function networks provides a more flexible model and gives a very good performance over a wide range of applications. However, in the modeling process, care is taken not to choose the number of the basis functions and the positions of the centres, the regularization parameter and the smoothing parameter as appropriate according to the...

In this paper, we propose a new method of projection pursuit, relative projection pursuit (RPP), which finds ‘interesting‘ low dimensional spaces different from reference data sets predefined by the user. In addition, as an application of the method, we develop a new dimension reduction method: sliced inverse regression with relative projection pur...

In this paper, we propose a new algorithm for Sliced Inverse Regression (SIR). SIR is a model for dimension-reduction of explanatory variables. There are several algorithms for the SIR model: SIR I, SIR II, etc. We have proposed an algorithm named SIRpp (SIR with Projection Pursuit). They find out a set of linear combinations of explanatory variabl...

We have to analyze enormous data in many cases. A personal computer can handle them, however, it would take a lot of time even if today's personal computer would have good specifications. Anyway, we have to seek a faster analysis environment. A parallel computer which has large computing power will satisfy us. Parallel Virtual Machine (PVM) is one...

In this paper, Sliced Inverse Regression (SIR) model is discussed from the viewpoint of distribution theory. Especially, we deal with conditional distribution of explanatory variables when the value of response variable is fixed. The concept of SIR is proposed by Li (1991) and the purpose of SIR is to reduce the dimension of explanatory variables....

In the paper, we investigate a conditional density function of sliced response variables and propose an algorithm for the sliced inverse regression (SIR) model with projection pursuit. The SIR model is a general model for dimension reduction of explanatory variables on regression analysis. Some algorithms for SIR model are proposed; SIR, SIR2, Biva...

1993年,1994年および1995年日本の夏期の天気は,極端な特徴をもって異なる傾向を示した.これら各年の夏期の天気図パターンの特徴を把握するために,主要点解析法(Principal Points Analysis)を組み込んだ多変量解析の適用を行った.まず,各年の7月と8月の日々の極東天気図パターンに対して40次元(高層天気図の場合は42次元)ベクトルによる数量化を行った.次に主成分分析の適用による次元の縮約および成分の直交化を行い,最後に主要点解析を適用した.その結果,地上天気図に関しては5個,高層天気図では4個の代表パターンが抽出された.さらに,日々の天気図パターンを最も類似度の高い代表パターンに類別し,年別の出現頻度を比較することによって,各夏期における特徴的な天気図パターンを特定...

In the present article, we discuss some conditions on the existence of asymmetric 2-principal points of univariate symmetric distributions and show some examples of 2-principal points that are asymmetric. k-Principal points of a distribution, proposed by Flury(1990), mean k points that minimize the expected squared distance from the nearest of the...

Linear regression or smoothing techniques are not adequate for curve fitting, in cases in which neither variable can be designated as the response. We present a new method for fitting bounded algebraic curve to multidimensional data using the least squares distance between data points and the curve. Numerical examples of the proposed method are als...

Principal Pointsは,Flury(1990)により提案され,クラスター分析におけるk-means法と同様の規準に基づいて,与えられた確率分布の密度関数をk個の領域に分割する際に最適となるような各領域のある種の中心点として求められる.このPrincipal Pointsは,理論的に様々な興味深い性質をもつほか,応用的にもクラスター分析や最適配置の理論などと関連が深く,天気図の解析に応用する研究も行われている.本論文では,対称性を有する確率分布が与えられた場合のPrincipal Pointsに関する理論的考察および計算機シミュレーションを行った.その結果,対称な1変量確率分布のうちロジスティック分布と両側指数分布に関しては3-Principal Pointsが期待値に関し対称であ...

In this paper we discuss projection pursuit into three dimensions. In the previous works, practically one- and two-dimensional projection pursuit have been dealt to find interesting structures of data. We identify structures of data with three-dimensional projection pursuit, which can not find with lower dimensional projection pursuit. The Friedman...

Most statistical softwares utilize graphical facility to display data mainly, not to construct and execute analysis. We had developed the data analysis system with visual manipulation and improve it on the UNIX platform with Tcl/Tk, the interface builder available on many architectures and operation systems. We offer its features and introduce some...

We deal with a method of fitting an algebraic curve and surface to multidimensional data without an external criterion. It is often sensible to treat one of the variables as a response variable and the other as an explanatory variable in other words data with an external criterion. The linear regression line or regression curve minimizes the sum of...

This paper presents a new method for fitting algebraic curves to multidimensional data using the exact squares distance between data points and the curve. Fitting smooth curves is one of the most important themes in pattern recognition and data analysis. Simple regression analysis or multivariate regression analysis are in use for a data set consis...

In order to simulate the large variation in tolerance doses for very small treatment volumes, we introduce a model which assumes the presence of cells which have migrated from unirradiated tissues. In order to represent serial architecture, the new model adds a new parameter to the familiar expression for serial architecture. Data derived from the...

We review graphical methods that are developed and/or improved in Japan. For example a decision of the bin-width on histogram with AIC, constellation graphical method, sampling method and a few methods with dynamic graphics are included. The computing environments for statistics in Japan are introduced in relation to these methods. In addition to t...

Most data analysis systems seem to be convenient for statisticians, but novices in statistics cannot master them easily since it is hard to select appropriate procedures for an observation. The application with the technique of knowledge processing to data analysis is an attractive idea, but there is little research on selecting procedures because...

A stability of clusters can be defined by the tolerance for a small disturbance in the data. An estimating expression of the stability is shown in reference to the case of single-linkage algorithm. As one of the applications of this stability, the method of deciding on the number of single-linkage clusters is proposed. Some numerical examples are s...

Generalized principal components analysis (GPCA) has been proposed by R. Gnanadesikan and M. B. Wilk [Data analytic methods in multivariate statistical analysis, in: P. R. Krishnaiah (ed.), Multivariate Analysis II (1969; Zbl 0212.220), 593-638] in order to analyze data structures by obtaining fitting functions to data points. The present paper tre...

A graphical representation for three-way data in a three-dimensional Euclidean space is presented in this paper. The method is an extension of Wakimoto and Taguri's Constellation Graphical Method; the projection of the three-dimensional graph on a plane is a constellation graph. Usage of the representation in a three-dimensional space is explained...

In this chapter, we deal with functional principal points and functional cluster analysis. The k principal points [4] are defined as the set of k points which minimizes the sum of expected squared distances from every points in the distribution to the nearest points of the set, and are mathematically equivalent to centers of gravity by k-means clus...

We propose a method for finding the algebraic curve that fits multi-dimensional data. An algebraic curve in n dimensional space is generally defined by n-1 polynomial expressions. The proposed method finds the n-1 polynomial expressions of the algebraic curve for n dimensional data. The sum of the squares' distances from the data point to the neare...

This article discusses the problem of choosing a regularization parameter in the group Lasso proposed by Yuan and Lin (2006), an l 1 -regularization approach for producing a block-wise sparse model that has been attracted a lot of interests in statistics, machine learning, and data mining. It is important to choose an appropriate regularization par...