Updated on 2024/02/28

写真a

 
NAGATA, Yasushi
 
Affiliation
Faculty of Science and Engineering, School of Creative Science and Engineering
Job title
Professor
Degree
工学博士 ( 大阪大学 )
(BLANK)

Education Background

  • 1982.04
    -
    1985.09

    Osaka University   Graduate School, Division of Engineering Science  

  • 1980.04
    -
    1982.03

    Osaka University   Graduate School, Division of Engineering Science  

  • 1976.04
    -
    1980.03

    Kyoto University   Faculty of Science   Mathematics  

Committee Memberships

  • 2023.11
    -
     

    日本品質管理学会  顧問

  • 2021.11
    -
    2023.11

    日本品質管理学会  会長

  • 2019.11
    -
    2021.11

    日本品質管理学会  理事,副会長

  • 2020.05
    -
     

    応用統計学会  監事

  • 2018.05
    -
    2020.05

    応用統計学会  理事

  • 2016.05
    -
    2018.05

    応用統計学会  会長

  • 2015.11
    -
    2017.11

    日本品質管理学会  理事,学会誌編集委員長

  • 2012.06
    -
    2016.05

    応用統計学会  編集委員

  • 2013.11
    -
    2015.11

    日本品質管理学会  監事

  • 2012.06
    -
    2015.06

    行動計量学会  編集委員

  • 2014.06
    -
     

    応用統計学会  理事

  • 2014.06
    -
     

    応用統計学会  評議員

  • 2014.06
     
     

    応用統計学会  評議員

  • 2012.06
    -
    2014.06

    応用統計学会  評議員

  • 2011.11
    -
    2013.11

    日本品質管理学会  理事

  • 2008.06
    -
    2012.06

    応用統計学会  編集理事

  • 2008.06
    -
    2010.06

    応用統計学会  評議員

  • 2007.10
    -
     

    日本品質管理学会  編集委員

  • 2005.10
    -
    2007.10

    日本品質管理学会  理事,論文誌編集委員長

  • 2005.10
    -
    2007.10

    日本品質管理学会  理事

  • 2001.10
    -
    2003.10

    日本品質管理学会  理事,学会誌編集委員長

  • 2000.10
    -
    2003.10

    日本品質管理学会  理事

  • 2000.06
    -
    2002.06

    応用統計学会  評議員

  • 1999.10
    -
    2001.10

    日本品質管理学会  編集委員

  • 1998.07
    -
    2000.07

    日本統計学会  理事

  • 1994.04
    -
    2000.03

    応用統計学会  編集委員

  • 1994.10
    -
    1997.09

    日本品質管理学会  評議員

  • 1995.01
    -
     

    日本科学技術連盟  デミング賞委員

  • 1992.07
    -
    1994.07

    日本統計学会  評議員

▼display all

Professional Memberships

  •  
     
     

    品質工学会

  •  
     
     

    日本品質管理学会

  •  
     
     

    日本統計学会

  •  
     
     

    日本行動計量学会

  •  
     
     

    日本経営工学会

  •  
     
     

    応用統計学会

▼display all

Research Areas

  • Statistical science

Awards

  • Best Paper Award

    2023.10   ANQ (Asian Network for Quality) Congress 2023  

  • 早稲田大学ティーチングアワード(2022年度春学期)

    2023.02   早稲田大学  

  • 日経品質管理文献賞

    2022.11  

  • Best Paper Award

    2022.10   ANQ (Asian Network for Quality) Congress 2022  

  • 早稲田大学ティーチングアワード総長賞(2021年度春学期)

    2022.02   早稲田大学  

  • Best Paper Award

    2021.10   ANQ (Asian Network for Quality) Congress 2021  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2020

    2020.10  

  • The Deming Prize for Individuals

    2019.11  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2019

    2019.10  

  • Waseda University Teaching Award (Spring Semester 2018)

    2019.02  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2017

    2017.09  

  • QMOD-ICQSS Conference, Best Paper Award Nomination

    2017.08  

  • The Nikkei QC Literature Prize

    2016.11  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2016

    2016.09  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2014

    2015.09  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2014

    2014.08  

  • 日経品質管理文献賞

    2013.11  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2013(2件受賞)

    2013.10  

  • 日経品質管理文献賞

    2012.11  

  • 日本品質管理学会最優秀論文賞

    2012.10  

  • Best Paper Award,ANQ(Asian Network for Quality) Congress 2012

    2012.08  

  • 日経品質管理文献賞

    2010.11  

  • 日経品質管理文献賞

    2009.11  

  • クオリティマネジメント賞

    2008.11  

  • 日本品質管理学会最優秀論文賞

    2004.10  

  • 日経品質管理文献賞

    1999.11  

  • 日本品質管理学会論文奨励賞

    1991.10  

▼display all

 

Papers

  • Ensemble Pruning of MT Method

    Katsuhiko Tateishi, Hiroki Iwamoto, Yasushi Nagata

    Total Quality Science   9 ( 1 ) 8 - 17  2023.10  [Refereed]

    Authorship:Last author

    DOI

  • The T-method with the application of sparse modeling

    Ryo Asano, Masato Ohkubo, Shinto Eguchi, Yasushi Nagata

    Total Quality Science   9 ( 1 ) 1 - 7  2023.10  [Refereed]

    Authorship:Last author

    DOI

  • A Consideration of the Recognition Taguchi Method Using High-Dimensional Principal Component Analysis

    Ryo Asano, Masato Ohkubo, Yasushi Nagata

    Total Quality Science   8 ( 2 ) 70 - 76  2023.06

    DOI

  • Adaptation of the Mahalanobis–Taguchi Method to Ordinal Scale

    Hiroki IWAMOTO, Yasushi NAGATA

    Total Quality Science   8 ( 1 ) 14 - 22  2022.12

    DOI

  • Mahalanobis-Taguchi Method for Anomaly Detection and Classification

    Kentaro Honma, Masato Ohkubo, Shinto Eguchi, Yasushi Nagata

    Total Quality Science   8 ( 1 ) 1 - 13  2022.12

    Authorship:Last author

    DOI

  • More Accurate Knowledge Search in Technological Development for Robust Parameter Design

    Kosuke Oyama, Masato Ohkubo, Yasushi Nagata

    Quality Innovation Prosperity   26 ( 1 ) 38 - 51  2022.03  [Refereed]

    Authorship:Last author

     View Summary

    Purpose: The causality search Taguchi (CS-T) method was proposed to support system selection in a robust parameter design. However, the target of the analysis is likely to be quasi-experimental data. This can be difficult to analyse with the CS-T method. Therefore, this study proposes a new analysis approach that can perform a more accurate knowledge search by applying the instrumental variable.Methodology/Approach: Using the CS-T method, appropriate knowledge search is difficult with quasi-experimental data, including endogeneity. We examined an analytical process that addresses the endogeneity between mechanism and output by utilizing the control and noise factors that constitute the mechanism as instrumental variables.Findings: The results show that 1) the proposed method has sufficient practical accuracy, even for quasi-experimental data including endogeneity; and 2) the extracted mechanism is less likely to fluctuate depending on the number of experimental conditions used. Moreover, we can clarify the position of the CS-T and proposed methods in system selection.Research Limitation/Implication: We perform estimation under the assumption that the threshold is known. However, the extracted mechanism may change depending on the threshold; this requires discussing how to determine them.Originality/Value of paper: Technological development requires a high degree of engineer sophistication. However, this study’s analytical process allows conducting more accurate knowledge search in a realistic and systematic way without requiring a high level of engineer input.

    DOI

  • Statistical analysis and prediction in American football

    Yu Takezaki, Yasushi Nagata

    Total Quality Science   7 ( 2 ) 51 - 59  2022.02  [Refereed]

    Authorship:Last author

    DOI

  • Basic Concepts of Multiple Comparison Procedures

    Yasushi Nagata

    Ouyou toukeigaku   51 ( 1-2 ) 31 - 42  2022  [Refereed]

    Authorship:Lead author

    DOI

  • A comparison of MT and RT methods for missing data analysis methods

    Kakeru Nishizawa, Masahiro Maeda, Yasushi Nagata

    Total Quality Science   7 ( 1 ) 42 - 50  2021.12  [Refereed]

    Authorship:Last author

    DOI

  • A study of MT systems applied to multivariate regression

    Shota Nakayama, Suguru Sekine, Yasushi Nagata

    Total Quality Science   7 ( 1 ) 10 - 22  2021.12  [Refereed]

    Authorship:Last author

    DOI

  • Setting Control Limits in a Multivariate Exponentially Weighted Moving Likelihood Control Chart

    Takumi SARUHASHI, Masato OHKUBO, Yuma UENO, Yasushi NAGATA

    Total Quality Science   6 ( 3 ) 84 - 90  2021.10  [Refereed]

    Authorship:Last author

    DOI

  • Application of VBGMM for pitch type classification: analysis of TrackMan's pitch tracking data

    Kazuhiro Umemura, Toshimasa Yanai, Yasushi Nagata

    Japanese Journal of Statistics and Data Science   4 ( 1 ) 41 - 71  2021.07  [Refereed]

    Authorship:Last author

     View Summary

    <title>Abstract</title>In the game of baseball, each pitcher throws various types of pitches, such as cutter, curve ball, slider, and splitter. Although the type of a given pitch may be inferred by audience and/or obtained from the TrackMan data, the actual pitch type (i.e., the pitch type declared by the pitcher) may not be known. Classification of pitch types is a challenging task, as pitched baseballs may have different kinematic characteristics across pitchers even if the self-declared pitch types are the same. In addition, there is a possibility that the kinematic characteristics of pitched baseballs are identical even if the self-declared pitch types are different. In this study, we aimed to classify TrackMan data of pitched baseballs into pitch types by applying the Variational Bayesian Gaussian Mixture Models technique. We also aimed to analyze the kinematic characteristics of the classified pitch types and indices related to batting performance while pitching each pitch type. The results showed that the pitch types could not be accurately classified solely by kinematic characteristics, but with consideration of the characteristics of the fastball the accuracy improves substantially. This study could provide a basis for the development of a more accurate automatic pitch type classification system.

    DOI

    Scopus

    3
    Citation
    (Scopus)
  • Study on Likelihood-Ratio-Based Multivariate EWMA Control Chart Using Lasso

    Takumi Saruhashi, Masato Ohkubo, Yasushi Nagata

    Quality Innovation Prosperity   25 ( 1 ) 3 - 15  2021.03  [Refereed]

     View Summary

    Purpose: When applying exponentially weighted moving average (EWMA) multivariate control charts to multivariate statistical process control, in many cases, only some elements of the controlled parameters change. In such situations, control charts applying Lasso are useful. This study proposes a novel multivariate control chart that assumes that only a few elements of the controlled parameters change.
    Methodology/Approach: We applied Lasso to the conventional likelihood ratio-based EWMA chart; specifically, we considered a multivariate control chart based on a log-likelihood ratio with sparse estimators of the mean vector and variance-covariance matrix.
    Findings: The results show that 1) it is possible to identify which elements have changed by confirming each sparse estimated parameter, and 2) the proposed procedure outperforms the conventional likelihood ratio-based EWMA chart regardless of the number of parameter elements that change.
    Research Limitation/Implication: We perform sparse estimation under the assumption that the regularization parameters are known. However, the regularization parameters are often unknown in real life; therefore, it is necessary to discuss how to determine them.
    Originality/Value of paper: The study provides a natural extension of the conventional likelihood ratio-based EWMA chart to improve interpretability and detection accuracy. Our procedure is expected to solve challenges created by changes in a few elements of the population mean vector and population variance-covariance matrix.

    DOI

  • Anomaly Detection for Noisy Data with the Mahalanobis–Taguchi System

    Masato Ohkubo, Yasushi Nagata

    Quality Innovation Prosperity   24 ( 2 ) 75 - 75  2020.07

    Authorship:Last author

     View Summary

    &lt;p&gt;&lt;strong&gt;Purpose:&lt;/strong&gt; Condition-based maintenance requires an accurate detection of unknown yet-to-have-occurred anomalies and the establishment of anomaly detection procedure for sensor data is urgently needed. Sensor data are noisy, and a conventional analysis cannot always be conducted appropriately. An anomaly detection procedure for noisy data was therefore developed.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methodology/Approach:&lt;/strong&gt; In a conventional Mahalanobis–Taguchi method, appropriate anomaly detection is difficult with noisy data. Herein, the following is applied: 1) estimation of a statistical model considering noise, 2) its application to anomaly detection, and 3) development of a corresponding analysis framework.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Findings:&lt;/strong&gt; Engineers can conduct anomaly detection through the measurement and accumulation, analysis, and feedback of data. Especially, the two-step estimation of the statistical model in the analysis stage helps because it bridges technical knowledge and advanced anomaly detection.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Research Limitation/implication:&lt;/strong&gt; A novel data-utilisation design regarding the acquired quality is provided. Sensor-collected big data are generally noisy. By contrast, data targeted through conventional statistical quality control are small but the noise is controlled. Thus various findings for quality acquisition can be obtained. A framework for data analysis using big and small data is provided.&lt;/p&gt;&lt;strong&gt;Originality/Value of paper:&lt;/strong&gt; The proposed statistical anomaly detection procedure for noisy data will improve of the feasibility of new services such as condition-based maintenance of equipment using sensor data.

    DOI

  • Proposal for a new single imputation method using Taguchi’s T-method

    Yuto Nakao, Yasushi Nagata

    Total Quality Science   5 ( 3 ) 102 - 110  2020.03

    Authorship:Last author

    DOI

  • Imbalanced data classification procedure based on SMOTE

    Daiki Gyoten, Masato Ohkubo, Yasushi Nagata

    Total Quality Science   5 ( 2 ) 64 - 71  2020.01  [Refereed]

    Authorship:Last author

    DOI

  • Anomaly detection for unlabelled unit space using the Mahalanobis–Taguchi system

    Masato Okubo, Yasushi Nagata

    Total Quality Management & Business Excellence   29 ( 5-6 ) 1213 - 1227  2020  [Refereed]

    Authorship:Last author

    DOI

    Scopus

    5
    Citation
    (Scopus)
  • Proposal for a univariate time series analysis method based on Taguchi's T-method

    Shushiro Matsui, Yasushi Nagata

    Total Quality Science   Vol.5 ( No.1 ) 1 - 10  2019.10  [Refereed]

    Authorship:Last author

    DOI

  • A Study on the Swiss Draw Style Tournament System Reflecting Game Results and Their Content

    Kazuhiro Umemura, Tomomichi Suzuki, Yasushi Nagata

    Total Quality Science   Vol.5 ( No.1 ) 45 - 52  2019.10  [Refereed]

    DOI

  • Conditional anomaly detection based on a latent class model

    Masato Ohkubo, Yasushi Nagata

    Total Quality Management &amp; Business Excellence   30 ( sup1 ) S227 - S239  2019.09  [Refereed]

    Authorship:Last author

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • 品質管理のための統計的方法

    永田靖

    プラスティック成形加工学会誌.   第31巻   132 - 136  2019.03  [Invited]

  • Estimation of the Effects of Parent-child Relationships on Children’s Academic Achievement

    Ayano KANBASHI, Masato OHKUBO, Yasushi NAGATA

    Kodo Keiryogaku (The Japanese Journal of Behaviormetrics)   46 ( 1 ) 1 - 10  2019  [Refereed]

    Authorship:Last author

    DOI

  • Analysis of data including missing values in the Taguchi’s T method

    Yuto Nakao, Yasushi Nagata

    Total Quality Science   4 ( 2 ) 53 - 64  2018.12  [Refereed]

    DOI

  • Application of AdaBoost to Taguchi’s MT Method

    Suguru Sekine, Yasushi Nagata

    Total Quality Science   4 ( 2 ) 65 - 74  2018.12  [Refereed]

    DOI

  • Using change point identification in financial data to detect turning points in companies&lt;/b&gt;

    Satoshi Kawamata, Yasushi Nagata

    Total Quality Science   4 ( 1 ) 41 - 52  2018.08  [Refereed]

    DOI

  • ロバスト・パラメータ設計を用いた最適株式ポートフォリオ選択

    神山紘樹, 永田 靖

    日本品質管理学会誌   48 ( 3 ) 64 - 75  2018.07  [Refereed]

    Authorship:Last author

    DOI

  • 一対比較法を順位推定に用いたスイスドロー式トーナメントの提案

    梅村和弘, 永田靖

    日本経営工学会論文誌   69 ( 1 ) 21 - 32  2018.04  [Refereed]

  • QUALITY DESIGN BASED ON TAGUCHI’S LOSS FUNCTION

    Masato Okubo, Yasushi Nagata

    , International Journal of Business Research   Vol.29   5 - 14  2018  [Refereed]

  • Anomaly detection in high-dimensional data with the Mahalanobis–Taguchi system

    Masato Ohkubo, Yasushi Nagata

    Total Quality Management & Business Excellence   29   5 - 14  2018  [Refereed]

    DOI

  • A proposal of regression hybrid modeling for combining random forest and x-means methods

    Yuma Ueno, Yasushi Nagata

    Total Quality Science   3 ( 1 ) 1 - 10  2017.08  [Refereed]

     View Summary

    <p>To derive useful information from complicated data, many hybrid modeling strategies that combine nonparametric and parametric methods have been proposed. In this study, we propose a new hybrid modeling strategy that combines the random forest and the x-means methods using linear regression analysis. This strategy is referred to as XR regression.This study has three purposes: to improve the performance of a strategy of hybrid modeling using the random forest method, to determine an optimal class automatically using the x-means method, and to compare the prediction accuracy of this method with that of other existing methods.To determine the characteristics of XR regression, we compare its prediction accuracy with that of the existing methods using Monte Carlo simulations.The simulation results show that XR regression has a high performance in any situation, especially in data sets that include interaction effects.</p>

    DOI CiNii

  • The Mahalanobis-Taguchi Method Based on Graphical Modeling

    Masato Ohkubo, Yasushi Nagata

    Japanese Journal of Applied Statistics   46 ( 1 ) 13 - 26  2017  [Refereed]

    Authorship:Last author

    DOI CiNii

  • Study of model evaluation method and of selection method of tuning parameter in Lass

    Tatsuki Inoue, Yasushi Nagata

    Total Quality Science   Vol.2 ( 1 ) 27 - 35  2016.07  [Refereed]

     View Summary

    <p></p><p>High-dimensional data sets have been used extensively in recent years. However, their use is computationally expensive, and model features are difficult to capture. Thus, it is necessary to construct a simple and appropriate model. The least absolute shrinkage and selection operator (Lasso) can estimate several regression coefficients as exactly zero, making it possible to estimate a model and select variables at the same time. However, several problems exist. The decision procedure for the tuning parameter, which is the coefficient of the penalty term, is not established, and it is difficult to determine the value of the estimated model.</p><p>In this paper, we consider a model evaluation method and a selection method for the tuning parameter. We investigate selection methods for the most appropriate tuning parameter and model evaluation methods using the model evaluation indexes RSS, AIC, BIC, and Mallows's Cp. Let several regression coefficients be zero and vary the number of zeroes. Then, we observe the performances of the evaluation indexes. Furthermore, we observe their performances when the values of the regression coefficients are close to zero and confirm them by simulation.</p>

    DOI CiNii

  • 連載講座「統計的方法の考え方を学ぶ」,第6回 実験計画法の考え方

    永田靖

    QCサークル誌     54 - 61  2016.06

  • Detection of change points in panel data based on Bayesian MT method

    Takumi Enomoto, Yasushi Nagata

    Total Quality Science   Vol.2 ( 1 ) 36 - 47  2016.06  [Refereed]

     View Summary

    <p></p><p>Panel data are a sort of multi-dimensional time-series data that consist of several sets of the observation unit as an identical community over time. Panel data include more information than time-series data or cross-section data. Using panel data, we are able to estimate more accurately, increase degrees of freedom, and avoid multi-collinearity.</p><p></p><p>In this paper, we consider a change point problem in panel data. Detection methods for change points in panel data have not been extensively proposed. This paper proposes a detection method for change points in panel data using the Mahalanobis–Taguchi (MT) method (Taguchi (2002)). As the MT method does not assume panel data or time-series data, we must extend the theory of the MT method. Furthermore, when the sample size is small, it is difficult to estimate the covariance matrix precisely (Miyakawa et al. (2007)). In this paper, the MT method is extended using Bayesian inference to resolve the above-mentioned difficulties. Conducting Monte Carlo simulations and a real data analysis, we show that our proposed method is useful.</p>

    DOI CiNii

  • 連載講座「統計的方法の考え方を学ぶ」,第5回 推定と検定の考え方

    永田靖

    QCサークル誌     54 - 61  2016.05

  • 連載講座「統計的方法の考え方を学ぶ」,第4回 時間的な変化を調べる

    永田靖

    QCサークル誌     52 - 59  2016.04

  • 連載講座「統計的方法の考え方を学ぶ」,第3回 母集団を理解する

    永田靖

    QCサークル誌     52 - 59  2016.03

  • 連載講座「統計的方法の考え方を学ぶ」,第2回 2次元でデータを見ることの効用

    永田靖

    QCサークル誌     56 - 63  2016.02

  • 連載講座「統計的方法の考え方を学ぶ」,第1回 不良と異常の違い

    永田靖

    QCサークル誌     52 - 59  2016.01

  • Robust parameter design with covariates of multiple noise factors

    Kensuke Goto, Hironao Sato, Masami Miyakawa, Yasushi Nagata

    Total Quality Science   Vol.1 ( 2 ) 77 - 88  2015.06  [Refereed]

     View Summary

    The intent of robust parameter design is to make a system insensitive to noise factors by choosing optimum levels for the controllable factors. This is conducted by finding the interactions between noise factors and control factors. In general, it is necessary to control the levels of noise factors; however, there are some that cannot be controlled. On the other hand, there are some which can be observed as covariates.<br/>Hirano and Miyakawa (2007) proposed a method based on linear regression to analyze the interactions between a single noise factor and the control factors, when the noise factor is observed as a covariate.<br/>This paper discusses robust parameter designs when multiple noise factors are observed as covariates. We propose two extensions of the Hirano and Miyakawa method: extended methods 1 and 2. We perform Monte Carlo simulations under several data models to estimate the accuracy of these methods. It is shown that the extended methods are able to analyze the interactions between each of the control factors and the noise factors, when multiple noise factors are observed as covariates. Since the noise factors are combined in extended method 1, it cannot detect which noise factor influences the control factors. On the other hand, extended method 2 distinguishes between noise factors, and thus it can detect the interactions between the control factors and each of the noise factors.

    DOI CiNii

  • タグチのT法における項目選択に関する研究

    河田紘志, 永田靖

    日本品質管理学会誌   45   179 - 193  2015.05  [Refereed]

    DOI

  • MTシステムにおける小標本データの解析方法

    大久保豪人, 永田靖

    日本経営工学会論文誌   66 ( 1 ) 30 - 38  2015.04  [Refereed]

     View Summary

    The Mahalanobis-Taguchi (MT) method is a standard method of multivariate analysis for detecting anomalies or recognizing patterns. A number of case studies using the MT method have been reported. However, good performance is only obtained when a sufficient number of samples can be ensured; if the number of samples is insufficient, this method has a large probability bias.  In this paper, we first analyze the existing measures of methods, in which performing dimension reduction, such as using variable selection, is common, and show that there are some problems with testing for unknown data. Secondly, we propose two analytical procedures for small sample data in which the detection capability with respect to unknown data is taken into account. In these proposed procedures, when the number of data samples is small compared to the dimensions of the variables, the detection measure in the MT method is replaced by a measure derived through approximating correlation matrices based on probabilistic principal component analysis (PPCA) or by introducing ensemble learning. Finally, based on raw data analysis using the KDDCup99 dataset and simulation results, we consider how the proposed procedures should be applied when multicollinearity occurs and which of these two procedures should be applied according to the data pattern.

    DOI CiNii

  • An application of a generalized inverse regression estimator to Taguchi&amp;apos;s T-Method

    Hiroshi Kawada, Yasushi Nagata

    Total Quality Science   1 ( 1 ) 12 - 21  2015

    Authorship:Last author

    DOI CiNii

  • Validation of and improvement to cyclic designs

    Yutaro Kaido, Yasushi Nagata

    Total Quality Science   1 ( 1 ) 32 - 40  2015

    Authorship:Last author

    DOI CiNii

  • Asymptotic Theory of Taguchi's Natural Estimators of the Signal to Noise Ratio for Dynamic Robust Parameter Design

    Koji Tsukuda, Yasushi Nagata

    COMMUNICATIONS IN STATISTICS-THEORY AND METHODS   44 ( 22 ) 4734 - 4741  2015  [Refereed]

     View Summary

    This article discusses the asymptotic theory of Taguchi's natural estimators of the signal to noise ratio (SNR) for dynamic robust parameter design. Three asymptotic properties are shown. First, two natural estimators of the population SNR are asymptotically equivalent. Second, both of these estimators are consistent. Finally, both of these estimators are asymptotically normally distributed.

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • 信号因子が複数存在するシステムに対する動特性ロバストパラメータ設計

    佃康司, 永田靖

    日本品質管理学会誌   44   83 - 91  2014.04  [Refereed]

    DOI

  • Several Properties of MT System and Improved Procedures

    Yasushi Nagata

    Japanese Journal of Applied Statistics   42 ( 3 ) 93 - 119  2013  [Refereed]

    DOI

  • タグチのRT法における同一次元でない連続量データへの適用方法

    大久保豪人, 永田靖

    日本品質管理学会誌   42 ( 2 ) 248 - 264  2012.04  [Refereed]

     View Summary

    Taguchi's RT method is designed under the assumption that all variables are measured in the same units. Therefore, an appropriate analysis cannot be necessarily done when applying RT method to continuous type data that the variables are measured in different units. This becomes a big problem when adapting RT method to various fields. In this paper, we present a new procedure to apply RT method to continuous type data that the variables have different units. And, which procedures becomes a powerful procedure is verified together with other procedures to which the full discussion has not been performed yet. Firstly, we evaluate the performance of all procedures by analyzing KDDCup99 dataset. Secondly, the feature of procedures that obtain good evaluations is compared and analyzed by simulation in detail. Our experiments with KDDCup99 dataset show that a new procedure has good performance. We also reveal that in a new procedure, it is useful for the analysis to evaluate variables more reasonably than other procedures, and to consider the correlation between variables. This study concludes that a new procedure is one of the powerful procedures and RT method can become a effective method to anomaly detection etc. for high-dimensional data by using it.

    DOI CiNii

  • タグチのT法およびその改良手法と重回帰分析の性能比較

    稲生淳紀, 永田靖, 堀田慶介, 森有紗

    日本品質管理学会誌   42 ( 2 ) 265 - 277  2012.04  [Refereed]

     View Summary

    Taguchi's T method belongs to the MT (Mahalanobis Taguchi) system. Several methods of MT system are used to detect outliers or extraordinary patterns, while the T method is used to predict future data. Since T method analyzes the same type of data set as the multiple regression analysis, the performance of T method has been compared with that of the multiple regression analysis under several case studies. This paper discusses the features of T method and considers two kinds of improved T methods. Furthermore, prediction accuracies of these methods are calculated by Monte Carlo simulations under several models. It is shown that the improved T methods are always superior to the original T method and that they are superior to the multiple regression analysis when the sample size is not large enough compared to the number of independent variables.

    DOI CiNii

  • 実験計画法のルーツ

    永田靖

    日本品質管理学会誌   41 ( 4 ) 441 - 445  2011.10

     View Summary

    This paper discusses the principles of experimental design proposed by R.A. Fisher. They are consisted of three principles which are replication, randomization and local control. This paper investigates the book "The Design of Experiments" written by R.A. Fisher in order to grasp what he considered and how he intended to describe these principles.

    DOI CiNii

  • 工程能力指数の点推定の改良について

    永田靖

    日本品質管理学会誌   41 ( 3 ) 392 - 398  2011.07  [Refereed]

     View Summary

    Process capability indices (PCI) are widely used in the activities of quality control. This paper discusses the improvement on the natural estimators for the basic PCI's, Cp, CpL, CpU and CpK. <br>Under the squared error loss function, Nagata (1995) showed that the natural estimator for Cp can be uniformly improved on by the estimator which is multiplied by some constant, and gave the degrees of improvement numerically. He also suggested by numerical calculations that the natural estimator for CpK cannot be improved on by the simple estimator which is multiplied by some constant.<br> In this paper it is shown that the natural estimator for CpU and CpL can be uniformly improved on by the estimators which are multiplied by some constant, and the degrees of improvement are given numerically. Furthermore, it is mathematically shown that the natural estimator for CpK cannot be improved on by any simple estimators which are multiplied by some constant when the sample size is larger than or equal to 7.

    DOI CiNii

  • 回帰分析と実験計画法

    永田靖

    日本食品機械研究会誌「食品加工技術」   30 ( 1 ) 20 - 26  2010.03

    CiNii

  • タグチのRT法で用いる距離の性質とその改良

    永田 靖, 土居大地

    日本品質管理学会誌   39 ( 3 ) 364 - 375  2009  [Refereed]

     View Summary

    RT (Recognition Taguchi) method is the newest among MT (Mahalanobis Taguchi) system. Several application cases have been published, while the properties of RT method have not been studied theoretically so far. The calculation is not complicated and RT method seems to be an attractive procedure with original ideas. However, practitioners should be cautious when they apply it to continuous type data. This paper discusses the following viewpoints. <BR> (1) The dimensions of all variables must be the same. Otherwise, one needs to translate the variables to non-dimension numbers. <BR> (2) The distance used in RT method has an essentially inappropriate property. It takes some large value at the center of the unit space and attains zero at the point of different position from the center. When a new member comes from the center of the unit space, it would be decided that the member is outside of the unit space, which leads to be misleading. <BR> (3) This paper proposes a new distance for RT system, which improves on the abovementioned inappropriate property. This new distance is zero at the center of the unit space, and then increases with variation as a member is getting further from the center.

    DOI CiNii

  • 実務に役立てるSQC教育とそのステップアップ

    永田 靖

    日本品質管理学会誌   39 ( 1 ) 7 - 14  2009.01

     View Summary

    It is discussed how to educate the practically useful SQC (statistical quality control). When a company holds an in-house training of SQC, there possibly exist several problems, which are the problem of selecting trainees, the problem of preparing adequate homework and examinations, the problem of engaging practical applications, the problem of evaluating the results and the problem of promoting a thoughtful secretary. In order to overcome these problems, a passionate support by the top is necessary and the existence of stars on SQC in the company is desired. As a guidepost for training a star, one particular topic is introduced, which is the way that one interprets the partial regression coefficients in regression equations. Furthermore, several topics which should be considered to use both the Taguchi methods and the SQC effectively are presented.

    DOI CiNii

  • 複数の測定箇所が存在する場合の工程能力指数

    飯島洋平, 永田 靖

    統計数理研究所共同研究リポート No.212 「極値理論の工学への応用(5)」     33 - 42  2008

  • 項目数がn-1以上の場合のMTシステムの第1種の距離

    永田 靖, 久冨 剛

    日本品質管理学会誌   38巻 ( 1 ) 142 - 146  2008  [Refereed]

     View Summary

    On Type I Distance of the MT System under the Condition that the Number of the Items is Larger than or Equal to n-1 Yasushi NAGATA/Go HISATOMI Under the condition that there are several linear restrictions (the condition 1) or that the sample size n is smaller than or equal to the number p of the items (the condition 2), the Mahalanobis distance cannot be calculated. In this situation it is called that there exists multicollinearity. Miyakawa and Nagata (Journal of Japanese Society Quality Control, Vol.33) derived two types of distances and recommended using them as a countermeasure against multicollinearity. This paper mathematically shows that the values of the type I distance are the same for the original data of the normal space when p≧n. As a result, the critical value cannot be determined based on the sampling distribution of the type I distance. It is essentially difficult to cope with the condition 2 in compared with the condition 1. Under the condition 2 some information has to be thrown away. This paper suggests the possibility of the methods. Furthermore, this paper also shows that the values of the Mahalanobis distance are the same for p=n-1 although there does not exist multicollinearity.

    DOI CiNii

  • タグチメソッドとSQCの“友好的”推進

    永田 靖

    クオリティマネジメント   58 ( 8 ) 10 - 17  2007.08

    CiNii

  • サンプルサイズの決め方

    永田靖

    医薬安全性研究会会報   50   1 - 46  2005

  • 「サンプルサイズの決め方」の講義における問の成績の分析

    永田靖

    医薬安全性研究会会報   50   47 - 50  2005

  • ダネットの方法におけるサンプルサイズの決め方について

    永田靖

    医薬安全性研究会会報   50   51 - 59  2005

  • 工程能力指数の正しい理解に向けて

    永田靖

    日本品質管理学会誌   34巻   133 - 136  2004

  • 企業におけるSQC教育への取組みの工夫と課題

    永田靖

    クオリティマネジメント   54,11月号 ( 11 ) 28 - 34  2003

    CiNii

  • A test of the equality of several SN ratios for the systems with dynamic characteristics

    Yasushi Nagata, Masami Miyakawa, Toshiyuki Yokozawa

    Journal of the Japanese Society for Quality Control   Vol.33   351 - 360  2003  [Refereed]

    DOI

  • マハラノビス・タグチ・システムにおける多重共線性対策について

    宮川雅巳, 永田靖

    日本品質管理学会誌   33巻 ( 4 ) 467 - 475  2003  [Refereed]

    DOI CiNii

  • On the Interpretatin of the Partial Regression Coefficient in Multiple Regression Analysis

    The Journal of the Japanese Society for Quality Control   32 ( 3 ) 125 - 136  2002.07  [Refereed]

    DOI

  • 連載「クオリティマネジメント講座」第4回科学的方法

    永田靖

    クオリティマネジメント   53巻,4月号 ( 4 ) 43 - 46  2002

    CiNii

  • TQCツールボックス「統計的方法」「多変量解析法」「実験計画法」

    永田靖

    日本品質管理学会誌   32巻   294 - 296  2002

  • 重回帰分析における偏回帰係数の解釈について

    仁科健, 永田靖

    日本品質管理学会誌   32巻   396 - 405  2002  [Refereed]

  • SQC教育で育む組織の風土

    永田靖

    日本品質管理学会誌   31巻   532 - 542  2001

  • 多重比較法としてのDuncan's multiple range testの適否

    永田靖

    日本医事新報   3967   116 - 117  2000

  • 工程能力指数Cpkの統計的推測

    永田靖

    福島大学経済学会商学論集   67巻 ( 3 ) 21 - 38  1999

    CiNii

  • 誤差分散のスタイン型改良信頼区間の最適性について

    永田靖

    岡山大学経済学会雑誌   30巻 ( 3 ) 767 - 776  1999

    CiNii

  • How to Use Multiple Comparison Procedures.

    Yasushi NAGATA

    Japanese journal of applied statistics   27 ( 2 ) 93 - 108  1998  [Refereed]

    DOI CiNii

  • Stein type confidence interval of the disturbance variance in a linear regression model with multivariate student-t distributed errors

    Yasushi Nagata

    Communications in Statistics - Theory and Methods   26 ( 2 ) 503 - 523  1997.01  [Refereed]

    DOI

    Scopus

    4
    Citation
    (Scopus)
  • 回帰分析における誤差分散のスタイン型改良信頼区間の諸性質

    永田靖

    岡山大学経済学会雑誌   28巻 ( 4 ) 1037 - 1060  1997

    CiNii

  • Optimality of the Stein‐type Estimator for Variance

    Yasuchi Nagata

    The Japanese Economic Review   48 ( 1 ) 113 - 117  1997.01  [Refereed]

    DOI

    Scopus

  • 回帰分析における母寄与率の推定量の比較

    永田靖

    日本品質管理学会誌   27巻   214 - 224  1997  [Refereed]

    DOI

  • The neyman accuracy and the wolfowitz accuracy of the stein type confidence interval for the disturbance variance

    Yasushi Nagata

    Communications in Statistics - Theory and Methods   25 ( 5 ) 985 - 1004  1996.01  [Refereed]

    DOI

    Scopus

    2
    Citation
    (Scopus)
  • Point estimation of the process capability index Cpk

    Yasushi Nagata

    岡山大学経済学会雑誌   26巻   663 - 677  1995

  • Fixed-width sequential confidence interval for the mean of a gamma distribution

    Yoshikazu Takada, Yasushi Nagata

    Journal of Statistical Planning and Inference   44 ( 3 ) 277 - 289  1995.01

    Authorship:Last author

    DOI

  • 2
    Citation
    (Scopus)
  • ON NORMAL APPROXIMATION TO THE DISTRIBUTION OF THE ESTIMATED PROCESS CAPABILITY INDEX _??_pk

    Yasushi Nagata

    JOURNAL OF THE JAPAN STATISTICAL SOCIETY   25 ( 2 ) 159 - 171  1995

    DOI CiNii

  • Confidence interval of process capability indices based on the discrete data

    Hidekazu Nagahata, Yasushi Nagata

    岡山大学教育学部研究集録   97巻   87 - 93  1994

    CiNii

  • Approximation formulas for the lower confidence limits of process capability indices

    Hidekazu Nagahata, Yasushi Nagata

    岡山大学経済学会雑誌   25巻   1121 - 1134  1994

  • An inequality in the condition for iproving on an equal-tails confidence interval of normal variance

    Taichi Inaba, Yasushi Nagata

    Journal of the Japan Statistical Society   Vol.24 ( 2 ) 181 - 184  1994  [Refereed]

    DOI CiNii

  • 工程能力指数の区間推定のすすめ

    永田靖

    品質管理   12月号 ( 12 ) 55 - 62  1993

    CiNii

  • Approximation formulas for the confidence intervals of process capability indices

    Yasushi Nagata, Hidekazu Nagahata

    Reports of Statistical Application Research, JUSE   Vol.39, No.3 ( 3 ) 15 - 29  1992

    CiNii

  • この手法・概念をこう教える-“推定”と“予測”の違い-

    永田靖

    日本品質管理学会誌   21巻   63 - 63  1991

  • 各手法の特徴比較(その3)

    永田靖

    統計数理研究所共同研究リポート27巻,「リスク評価のための毒性・薬効データ解析法の研究」     23 - 40  1991

  • 決定樹方式での統計手法の利用の注意点

    永田靖

    医薬安全性研究会会報   34巻   26 - 39  1991

  • 指数分布の尺度母数の改良信頼区間

    永田靖

    日本品質管理学会誌   21巻 ( 1 ) 5 - 10  1991  [Refereed]

     View Summary

    Let x(1), x(2), …, x(n) be the order statistics of a sample of size n from the exponential distribution with an unknown location-scale parameter (μ,σ). It is shown that usual confidence intervals for σ based on the only statistic Σ^^^n__<i=1>(x(i)-x(1)) can be improved by using the statistic x(1). Further the degree of the improvement is computed numerically.

    DOI CiNii

  • 工程能力指数の区間推定

    永田靖

    日本品質管理学会誌   21巻   109 - 114  1991  [Refereed]

     View Summary

    The process capability indices (CPI) are often used to obtain the information about the production process capability and there are some criteria in determining it, the criteria which are designed for the true values of CPI's. However, since the true values of CPI's are unknown in practice, they are estimated from several data. So the criteria might not be applied to these (point-) estimated values. Confidence intervals for CPI's should be constructed and these results should be taken into consideration. In this paper some useful and practical methods for the interval estimation are given.

    DOI CiNii

  • 多重比較における下降手順に対応する同時信頼区間の非妥当性についての考察

    永田靖, 松田眞一

    岡山大学経済学会雑誌   23巻 ( 3 ) 449 - 466  1991

    CiNii

  • 各手法の特徴比較(その2)

    永田靖

    統計数理研究所共同研究リポート22巻,「毒性・薬効データ解析における多重比較法の利用」     27 - 45  1990

    CiNii

  • 分散比の改良信頼区間とその性能について

    永田靖

    岡山大学経済学会雑誌   22巻 ( 2 ) 267 - 282  1990

    CiNii

  • Definition of Powers in Multiple Comparisons, and Features of Several Procedures Based on Them

    Shin-ichi MATSUDA, Yasushi NAGATA

    Japanese journal of applied statistics   19 ( 2 ) 93 - 113  1990  [Refereed]

    Authorship:Last author

    DOI CiNii

  • 各手法の特徴比較

    永田靖

    統計数理研究所共同研究リポート18巻,「多重比較方式の諸問題」     15 - 30  1989

  • 直交表におけるプーリングを伴った誤差分散の推定

    永田靖

    日本品質管理学会誌   19巻   12 - 19  1989  [Refereed]

    DOI CiNii

  • Improvements of interval estimations for the variance and the ratio of two variances

    Yasushi Nagata

    Journal of the Japan Statistical Society   Vol.19 ( 2 ) 151 - 161  1989  [Refereed]

    DOI CiNii

  • 実験計画法をめぐる諸問題-プーリング・逐次検定-

    永田靖

    日本品質管理学会誌   18巻   196 - 204  1988

  • Minimaxity and nonminimaxity of a preliminary test estimator for the multivariate normal mean

    Taichi Inaba, Yasushi Nagata

    Annals of the Institute of Statistical Mathematics   39 ( 1 ) 49 - 54  1987.12  [Refereed]

    DOI

  • 変量模型の分散分析における実際的なプーリング手法

    永田靖

    日本品質管理学会誌   17巻   276 - 282  1987  [Refereed]

    DOI

  • Unbiased variable selection procedure in regression models

    Yasushi Nagata

    Journal of the Japan Statistical Society   Vol.17 ( 2 ) 175 - 183  1987  [Refereed]

    DOI CiNii

  • Data-compatible model selection and a minimax family of model selection and estimation procedure for a multivariate normal mean

    Yasushi Nagata

    Journal of the Japan Statistical Society   Vol.16 ( 1 ) 67 - 73  1986  [Refereed]

    DOI CiNii

  • Minimaxity of a preliminary test estimator for the mean of normal distribution

    Yasushi Nagata, Taichi Inaba

    Annals of the Institute of Statistical Mathematics   37 ( 3 ) 437 - 442  1985.12  [Refereed]

    DOI

  • Interval estimation of a cell mean in a fixed ANOVA model under a preliminary test

    Yasushi Nagata, Takaharu Araki

    Reports of Statistical Application Research, JUSE   Vol.32, No.1   1 - 10  1985  [Refereed]

  • Practical pooling method in a fixed ANOVA model, size and power trade-off

    Yasushi Nagata, Takaharu Araki

    Reports of Statistical Application Research, JUSE   Vol.32, No.2 ( 2 ) 1 - 9  1985  [Refereed]

    DOI CiNii

  • Contribution to the theory of successive inference from a practical and a theoretical view point

    Yasushi Nagata

    Dissertation(Osaka University)    1985  [Refereed]

  • Minimaxity and inadmissibility of the model selection and estimation procedure for the mean of a multivariate normal distribution

    Yasushi Nagata

    Journal of the Japan Statistical Society   Vol.15 ( 2 ) 177 - 182  1985  [Refereed]

    DOI CiNii

  • An admissible estimator in the one-parameter exponential family with ambiguous information

    Yasushi Nagata

    Annals of the Institute of Statistical Mathematics   35 ( 2 ) 193 - 199  1983.12  [Refereed]

    DOI

  • Admissibility of some preliminary test estimators for the mean of normal distribution

    Yasushi Nagata

    Annals of the Institute of Statistical Mathematics   35 ( 3 ) 365 - 373  1983.12  [Refereed]

    DOI

  • Estimationn of the Pareto parameter and its admissibility

    Yasushi Nagata

    Mathematica Japonica   Vol.28   149 - 155  1983  [Refereed]

▼display all

Books and Other Publications

  • タグチメソッドの探究

    宮川雅巳, 永田靖

    日科技連出版社  2022.01

  • SQCセミナーの物語,2020品質月間テキストNo.449

    永田靖

    日本科学技術連盟  2020.10

  • 統計的方法の考え方を学ぶ

    永田靖

    日科技連出版社  2016

  • 開発・設計に必要な品質管理

    日本品質管理学会中部支部産学連携研究会, 田靖

    日本規格協会  2015.11

  • アンスコム的な数値例で学ぶ統計的方法23講

    廣野元久, 永田靖

    日科技連出版社  2013.01

  • 工程能力指数

    永田靖, 棟近雅彦

    日本規格協会  2011.11

  • 開発・設計における“Qの確保”

    日本品質管理学会中部支部産学連携研究会編

    日本規格協会  2010.05

  • データ解析に役立つExcel関数

    芳賀敏郎, 永田靖

    日科技連出版社  2009.12

  • 統計的品質管理

    永田靖

    朝倉書店  2009

  • 新版 品質保証ガイドブック

    社)日本品質管理学会編, 永田靖, 編集委員, 分担執筆

    日科技連出版社  2009

  • 統計データ科学事典

    杉山高一, 藤越康祝, 杉浦成昭, 国友直人編, 担執筆

    朝倉書店  2007

  • 日本規格協会品質管理と標準化セミナーテキスト「感性品質の評価」

    池山豊, 永田靖

    日本規格協会  2006

  • 日本規格協会品質管理と標準化セミナーテキスト「検定と推定」

    永田靖

    日本規格協会  2006

  • 日本規格協会品質管理と標準化セミナーテキスト「統計的方法の基礎」

    永田靖

    日本規格協会  2006

  • 日科技連品質管理セミナーベーシックコーステキスト第8章「分散分析」

    永田靖

    日本科学技術連盟  2006

  • 品質管理のための統計手法

    永田靖

    日本経済新聞社  2006

  • 統計学のための数学入門30講

    永田靖

    朝倉書店  2005

  • おはなし統計的方法

    永田靖

    日本規格協会  2005

  • クォリティマネジメント用語辞典

    吉澤正編, 永田靖, 編集委員, 分担執筆

    日本規格協会  2004

  • サンプルサイズの決め方

    永田靖

    朝倉書店  2003

  • 生産管理用語辞典

    永田靖, 分担執筆

    日本規格協会  2002

  • SQC教育改革

    永田靖

    日科技連出版社  2002

  • 多変量解析法入門

    永田靖, 棟近雅彦

    サイエンス社(中国語訳:陳耀茂訳,全華科技図書股イ分有限公司,2003.)  2001

  • 入門実験計画法

    永田靖

    日科技連出版社(中国語訳:陳耀茂訳,中衛發展中心,2004.)  2000

  • 日科技連品質管理セミナーベーシックコーステキスト第11章「実験計画法」

    永田靖, 棟近雅彦

    日本科学技術連盟  1999

  • グラフィカルモデリングの実際

    日本品質管理学会テクノメトリックス研究会編

    日科技連出版社  1999

  • 統計的多重比較法の基礎

    永田靖, 吉田道弘

    サイエンティスト社  1997

  • 統計的方法のしくみ

    永田靖

    日科技連出版社  1996

  • 日科技連品質管理セミナーベーシックコーステキスト第5章「計数値データの解析」

    永田靖

    日本科学技術連盟  1992

  • 入門統計解析法

    永田靖

    日科技連出版社  1992

▼display all

Research Projects

  • タグチメソッドの精緻化の研究

    日本学術振興会  科学研究費助成事業

    Project Year :

    2022.04
    -
    2025.03
     

    永田 靖

  • 複雑なデータ環境に適用できるタグチメソッドの研究

    Project Year :

    2018.04
    -
    2021.03
     

     View Summary

    SN比解析や田口流実験計画法に関連して次のような研究を行った.田口の導入したサイクリック計画を用いて,スポーツの対戦システムに関する最適性について研究した昨年の結果をさらに発展させて論文にまとめた.この結果は学術誌に掲載された.また,一般化線形モデルを用いた動特性に関するロバスト・パラメータ設計の研究を国際会議で発表し,ベスト・ペーパー・アワードを受賞した.さらに,多変量累積和管理図について両側信頼限界を設定する方法を開発し,その性能の検討した.この成果を国際会議で発表した.MTシステムに関連して,次のような研究を行った.まず,異常検知のためのMT法の研究について述べる.潜在クラスモデルに基づく条件付き異常検知システムを開発して,その性能が優れていることを確認した.この成果を国際会議で発表し,論文にまとめて学術誌に投稿し,掲載された.ラベルをつけていない単位空間を用いたMT法に基づく異常検知の方法を開発し,性能評価した.この成果を学術誌に論文を投稿し,採択された.次に,予測手法であるT法に関連した研究について述べる.まず,単一変数の時系列モデルのデータを変換して解析するT法タイプの方法を開発し,その性能評価についての研究を行った.この成果をまとめ,学術誌に論文を投稿し,掲載された.次に,T法を用いて欠測値補完を行う方法を開発し,性能評価を行った.その成果を学術誌に論文と投稿し,掲載が決定した.不均衡データに対してデータを増加させるスモートの新しい方法を提案し,分類問題に適用し,性能が優れていることを確認した.その成果を論文にまとめ,その改訂作業を行い,学術誌に掲載された.SN比解析や田口流実験計画法の研究ついては,ほぼ順調に進展した.MTシステムの研究についても,複雑な状況下を設定したもとで,十分多くの研究成果を出すことができた.また,多くの研究成果の論文化と掲載まで達成することができた.より複雑な状況におけるSN比解析や田口流実験計画法,そしてMTシステムの研究については研究を継続してきたい.特に,複雑な状況設定については,データの構造の複雑さ,変数の多次元化・高次元化,データの欠損や不均衡の状況に対して,研究を深めていきたい

  • Extensive research of Taguchi method

    Project Year :

    2015.04
    -
    2018.03
     

     View Summary

    In this research, we propose extended Taguchi methods and aim for the theoretical development of the extended methods. In particular, the generalization and extensional improvement of the SN ratio in the experimental design (hereafter referred to as research I) and the extended improvement of the MT system (Mahalanobis Taguchi system) and its theoretical research (hereinafter referred to as research Ⅱ).As for research Ⅰ, we will clarify the properties of the SN ratio proposed in various situations from the mathematical aspect, propose new analytical methods using the new SN ratios, and clarify their theoretical properties . As for research Ⅱ, we theoretically clarify the features of each method of MT system, develop new analytical methods, compare with existing methods among MT system and multivariate analysis methods. Furthermore, I want to formulate guidelines

  • Systematic theorization of Taguchi method

    Project Year :

    2012.04
    -
    2015.03
     

     View Summary

    Taguchi method is widely used in practice by engineers in the world. Taguchi method is the collection of procedures which Dr. Gen-ichi Taguchi had been developing for years.Particularly, Taguchi method consists of (i) robust parameter design based on SN ratio, (ii) MT system which is the collections of several multivariate data analysis procedures, and (iii) the evaluation method based on loss functions. In this research, we took up (i) and (ii). We proved some theoretical aspects of the existing procedures and then we proposed several improving procedures. Our results are rated highly, and we got several awards for our research

  • Theoretical studies on Taguchi methods and their applications

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research

    Project Year :

    2007
    -
    2009
     

    TANAKA Kentaro, NAGATA Yasushi, NAKANISHI Hiroko, SEKI Yoichi, YAMADA Syu, NISHINA Ken, AOKI Satoshi, KURUSHIMA Aiko, MIYAKAWA Masami

     View Summary

    In this research, we investigated the mathematical aspects of Taguchi method. Especially, we studied on the following four topics : i) statistical inference of SN ratio, ii) analysis of orthogonal array, iii) online quality control, iv) theoretical analysis of MT system. As a result, we obtained some new results : an improved version of T method, a new method for diagnosis of process by change-point detection, an application of MT system for outlier detection, a new method for supersaturated designs, and so on

  • Sample size determination for procedures of design of experiments and multivariate analysis

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research

    Project Year :

    2005
    -
    2007
     

    NAGATA Yasushi

     View Summary

    In this research, we dealt with multiple comparison procedures and Mahalanobis Taguchi system(MT system). We investigated their performance from the view points of sample size determination and proposed several new procedures.Firstly, we investigated the following topics about multiple comparison procedures and obtained the following results. (1) We reorganized the known results about sample size determination for Dunnett procedure and proposed an approximation method. (2) We examined the robustness of Max t procedure for underling assumptions and the relation to the sample size. (3) We examined the performance of maximum contrast procedure and also proposed some improved version of it. We think that these results will be useful for users to apply in practical situations and that they will keep users from misusing multiple comparison procedures.Secondly, we investigated the following topics about MT system and obtained the following results.(1) We establish the relation between the distribution of signal-to-noise ratio and doubly noncentral F distribution, which will be able to enhance the theoretical research of Taguchi methods.(2) We clarified the position of MT system in the Taguchi method.(3) We derived a theorem concerning some distance used in MT system.(4) We compared the performance of several procedures under the multicollinearity. These results contribute to the theoretical development of MT system.Furthermore, we investigated model selection methods for other multivariate analysis and obtained useful results

  • Theoretical Research on Graphical Modeling and New Multivariate Analysis and the Development of the Software

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research

    Project Year :

    1998
    -
    1999
     

    NAGATA Yasushi, TSUBAKI Hiroe, KURIKI Satoshi, KANO Yutaka, MIYAKAWA Masami, NISHINA Ken

     View Summary

    We studied the statistical multivariate analysis, especially the graphical modeling.The summary of theoretical research results is following : (1) multiplicity of tests and model selections ; (2) performance of deviance as a statistic for the validity of a model ; (3) a new variable selection procedure based on a path diagram ; (4) the effect of intervention on the independent variables using the directed model as a causal model ; (5) the performance of the conditional intervention ; (6) stepwise variable selection incovariance structure analysis ; (7) shrinkage estimation to smooth boundary ; (8) statistical aspect of process capability indices.The summary of the research results from practical view points is following : (1) the development and the improvement of a computer software for the graphical modeling of continuous variables ; (2) the development and the improvement of a computer software for the graphical modeling of the discrete wariables ; (3) the applied examples on the process control, remote sensing and customer satisfaction survey.We published the book "Introduction to graphical modeling" . This book was selected as the Nikkei Hinshitsukanri Bunkensho in 1999

  • DEVELOPMENT OF ANALYSIS SYSTEM FOR QUALITY INFORMATION BY GRAPHICAL MODELLING

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research

    Project Year :

    1996
    -
    1997
     

    MIYAKAWA Masami, NISHINA Ken, NAGATA Yasushi, TSUBAKI Hiroe, KURIKI Satoshi, IIDA Takahisa

     View Summary

    A graphical model is a probability model for random observations whose independence structure can be characterised by independence graph, and graphical modelling is the statistical activity of fitting graphical models to the data. The theory of graphical modelling has emerged from a mixture of log-linear models in multi-way contingency tables and covariance selection models in multivariate normal distribution. In this research, we develop the conversational data analysis system both for graphical Gaussian modelling which can be used for analysis of multi-dimensional continuous variables when assuming their distribution to be multivariate normal distribution, and for graphical log-linear modelling which can be used for multi-dimensional discrete variables when assuming their distribution to be mulinomial distribution. This system is applicable to representation of a probability model by using not only undirected independence graphs but also directed and chain independence graphs. The system can be linked to the existing analysis system for other methods of multivariate analysis

  • Statistical Models for Causal Inference

     View Summary

    In this research, statistical inference procedures of causal diagram using graphical modeling and covariance structure analysis are discussed. We formulate the causal interpretation and outline the proper use of structural equation model and potential response model. The randomization-based estimator of causal effect is given which is interpreted as an extension of the instrumental variables estimator. The inference problem of joint interventions in some treatment variables is also investigated. The parametric expressions for the joint intervention effects using the linear structural equation model are given

▼display all

Misc

 

Syllabus

▼display all

 

Sub-affiliation

  • Faculty of Science and Engineering   Graduate School of Creative Science and Engineering

Research Institute

  • 2023
    -
    2024

    Center for Data Science   Concurrent Researcher

  • 2022
    -
    2024

    Waseda Research Institute for Science and Engineering   Concurrent Researcher

Internal Special Research Projects

  • 統計的多重比較法の研究

    2004  

     View Summary

     次の6つの課題に取り組んだ.(1)検出力とサンプルサイズの検討 統計的方法の基礎的な手法についてサンプルサイズの設計の方法を,検出力と区間推定の立場からまとめ,成書とした.このようなまとまった書籍はこれまでになかったものである.(2)多重比較法におけるサンプルサイズの検討 多重比較法におけるサンプルサイズの設計方法について検討した.(3)工程能力指数の研究 工程能力指数について,データが正規分布に従っていない場合も含めて,推定方法や精度などについて研究した.(4)実験計画法における推定方法の改良について 実験計画法で得られるデータに基づいて,効果の推定や誤差分散の推定の方法について,従来方法の改良方法を開発し,その推定精度について研究した.(5)新しい多変量解析の適用についての研究 グラフィカルモデルを用いた因果推論について,回帰分析の枠組みを発展させる方向で研究した.また,マハラノビスタグチシステムについて,新しい尺度の構成方法などについて研究した.(6)統計学のための数学のまとめ 統計学を学ぶために必要な数学を吟味し,まとめた.

  • フィッシャー流実験計画法と田口流実験計画法との融合

    2002  

     View Summary

    次の8つの課題に取り組んだ.(1)検出力とサンプルサイズの検討 広く使われている標準的な統計的方法について検定における検出力と区間推定に基づく観点から,非心分布の各種の近似式を検討し,実務的に使える近似式を整理した.それに基づいて,サンプルサイズの設計方式を整理した.今後の教育・研究に活用するため,それらの結果を,200ページあまりの冊子にまとめた.特に,今後は,多重比較法について,検出力とサンプルサイズの設計の問題を取り組んでいきたい.今回の成果は,その基礎となるものである.(2)SN比についての検定 測定機の性能を測る量としてSN比が提案されている.本研究では,複数の測定器のSN比の一様性の検定方法を導き,検定のサイズと検出力について検討した.簡便でリーズナブルな精度をもつ検定方式を得ることができた.(3)重回帰分析における偏回帰係数の解釈について 重回帰分析における偏回帰係数の解釈の問題は困難である.これを,グラフィカルモデリングの観点からの解釈と関連させて整理した.(4)実験計画法における最適水準での推定量の統計的性能について 分散分析表に基づいて,最適水準を決定して母平均の推定量を構成する場合,最適水準の選択による上方へのバイアスと,効果があるのにそれを無視してしまうことによる下方へのバイアスが存在する.それらを適切に評価し,制御することが,精度のよい推定につながる.本研究では,これらの観点から,推定量のバイアスと平均2乗誤差を評価し,どのような推定方式がよいのかを検討した.(5)重回帰分析における寄与率の推定 重回帰分析では,寄与率や自由度調整済み寄与率は基本的な統計量であり,常に参照される量である.しかし,それらが真の寄与率を精度良く推定できているとは限らない.モデルを固定させたときに,これらの推定精度を研究した論文は存在したが,本研究では,変数選択を伴った状況で,これらの推定精度について研究した.(6)偏相関係数行列の感度分析 偏相関係数行列はグラフィカルモデリングにおいて重要である.本研究では,偏相関係数行列について,個々のデータがどのような影響を有するのかという観点から,感度分析の研究を行った.(7)MTA法についての研究 MTA法はタグチメソッドにおいて最近開発された方法である.通常のマハラノビスの距離のかわりに,余因子に基づいて距離を作成する.本研究では,その問題点を整理し,2つの軸により尺度を構成する方法を提案した.(8)実践的な統計教育の研究 企業における多変量解析法や実験計画法に関する統計教育の問題点を整理し,改良案を提案した.

  • 実験計画法の工程解析への応用

    2000  

     View Summary

     次の4つの内容に取り組んだ。(1)工程能力指数のロバスト推定 将来の「安定した母集団分布」を想定して、それとは異質の母集団分布からのデータが混在している状況を設定し、得られるデータに基づいて「安定した母集団分布」に基づく工程能力指数の推定の研究を行った。通常の最尤法に基づく推定方法が異質な母集団分布からのデータにより大きく影響を受けることを示し、次にそのようなデータからの影響が小さいロバストな推定方法を提案し,その数学的性質および定量的な数値評価を行った。(2)回帰分析における寄与率の推定 回帰分析における寄与率の点推定の方法を研究した。モデルを確定した下で寄与率を精度よく推定するための研究は従来から行われていたが、本研究では、モデル選択を行う過程を定式化し、その下で寄与率を推定する問題を研究した。自由度調整済寄与率の精度が比較的良好であるという結果を得た。 現在、既存の推定方法の比較検討の段階に留まっているが、今後、新たな推定量の提案と区間推定の問題、さらには、回帰分析を拡張したモデルに対しても取り組んでいきたい。(3)統計的多重比較法に関する研究 ダネットの方法およびそれを拡張した方法について研究した。今後、具体的な解析を行うための数値表などを整備する予定である。また、ダンカンの多重比較法についても検討した。(4)学部学生・大学院生向けのテキストの作成 学部学生や大学院学生がこの分野の研究に入って行くにあたって、効率的にその内容を学べるようなテキストを作成した。これらには、筆者が研究している内容も取り入れている。