TOHOKU UNIVERSITY Researchers

Details of the Researcher

Home

日本語 English

Yasumasa Matsuda

Section

Graduate School of Economics and Management

Job title

Professor

Degree

博士（理学） (Tokyo Institute of Technology)

researchmap

https://researchmap.jp/read0143857

J-GLOBAL ID

200901061245527028

Research History 4

2000/04 - 2005/09

新潟大学経済学部　助教授
2003/09 - 2004/09

Department of Statistics, London School of Economics　研究員
1997/04 - 2000/03

神奈川大学工学部　助手
2005/10 -

Faculty of Economics, Tohoku University　助教授

Education 2

Tokyo Institute of Technology　情報理工学研究科　department of mathematical computing sciences

- 1999/03
Tokyo Institute of Technology　Faculty of Science　department of mathematics

- 1991/03

Committee Memberships 4

Spatial Econometrics Association　Journal of Spatial Econometrics Association, Associate Editor

2020/04 - Present
American Mathematical Society　Reviewer

2005/04 - Present
Japanese Journal of Statistics and Data Science　Japanese Journal of Statistics and Data Science, Coodinating Editor

2018/06 - 2023/05
Japan Statistical Association　board member

2017/06 - 2021/05

Professional Memberships 3

日本統計学会
応用統計学会

2008/09 - Present
Bernoulli Society for Mathematical Statistics and Probability

Research Interests 6

ピリオドグラム
スペクトル
spectral analysis
spatial temporal data
spatial data
time series

Research Areas 2

Humanities & social sciences / Economic statistics /
Informatics / Statistical science /

Awards 3

JSPS prize

2011/03/03　日本学術振興会　時空間統計学の理論と空間計量経済学への応用
JSS Research Award

2019/09
応用統計学会賞

2008/06　応用統計学会

Papers 33

Spatial extension of generalized autoregressive conditional heteroskedasticity models Peer-reviewed

Takaki Sato, Yasumasa Matsuda

SPATIAL ECONOMIC ANALYSIS　16　(2)　148-160　2021/04

DOI： 10.1080/17421772.2020.1742929 　

ISSN： 1742-1772

eISSN： 1742-1780
Continuous auto-regressive moving average random fields on (n) International-journal Peer-reviewed

Peter J. Brockwell, Yasumasa Matsuda

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY　79　(3)　833-857　2017/06

DOI： 10.1111/rssb.12197 　

ISSN： 1369-7412

eISSN： 1467-9868
Fourier analysis of irregularly spaced data on R-d Peer-reviewed

Yasumasa Matsuda, Yoshihiro Yajima

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY　71　(1)　191-217　2009

DOI： 10.1111/j.1467-9868.2008.00685.x 　

ISSN： 1369-7412
A test statistic for graphical modelling of multivariate time series Peer-reviewed

Yasumasa Matsuda

BIOMETRIKA　93　(2)　399-409　2006/06

DOI： 10.1093/biomet/93.2.399 　

ISSN： 0006-3444

eISSN： 1464-3510
Series ridge regression for spatial data on R𝑑 Peer-reviewed

Daisuke Kurisu, Yasumasa Matsuda

Bernoulli　32　(1)　2026/02

DOI： 10.3150/25-BEJ1871 　
Density‐Valued ARMA Models by Spline Mixtures Invited Peer-reviewed

Yasumasa Matsuda, Rei Iwafuchi

Journal of Time Series Analysis　2025/10/14

DOI： 10.1111/jtsa.70025 　
How long do voluntary lockdowns keep people at home? The role of social capital during the COVID-19 pandemic

Yuta Kuroda, Takaki Sato, Yasumasa Matsuda

The Japanese Economic Review　2025/05/26
Publisher: Springer Science and Business Media LLC
DOI： 10.1007/s42973-025-00206-8 　

ISSN： 1352-4739

eISSN： 1468-5876

More details Close

Abstract Using data from over 80 million mobile devices, we create a city-by-day-level mobility index for the duration of the COVID-19 pandemic to analyze how voluntary prevention and policy compliance varies with social capital levels. We find that in the second year of the pandemic, both voluntary preventative activities and policy compliance were substantially reduced in areas with low levels of social capital but not in areas with high levels of social capital. Additionally, the impact of social capital is heterogeneous, and norms and culture may play a minor role in more developed and urbanized areas. Additionally, in Japan, the observed heterogeneity in behavior regarding supporting political parties was not based on ideology, position, or ruling status but rather on whether a party was a majority or minority. These results suggest that valuing conformity with others is an essential driver of behavior that benefits the community.
Stock Market Index Movement Prediction Using Partial Contextual Embedding BERT-LSTM

Youjia Liu, Yasumasa Matsuda, Zhijie Zhang

Lecture Notes in Computer Science　49-60　2024/11/12
Publisher: Springer Nature Singapore
DOI： 10.1007/978-981-96-0119-6_5 　

ISSN： 0302-9743

eISSN： 1611-3349
Local polynomial trend regression for spatial data on Rd

Daisuke Kurisu, Yasumasa Matsuda

Bernoulli　30　(4)　2024/11/01
Publisher: Bernoulli Society for Mathematical Statistics and Probability
DOI： 10.3150/23-bej1694 　

ISSN： 1350-7265
Spatial extension of mixed models of the analysis of variance

Takaki Sato, Yuta Kuroda, Yasumasa Matsuda

Spatial Economic Analysis　1-15　2024/02/26
Publisher: Informa UK Limited
DOI： 10.1080/17421772.2024.2312132 　

ISSN： 1742-1772

eISSN： 1742-1780
Estimation of Large Covariance Matrices with Mixed Factor Structures

Runyu Dai, Yoshimasa Uematsu, Yasumasa Matsuda

The Econometrics Journal　2023/09/27
Publisher: Oxford University Press (OUP)
DOI： 10.1093/ectj/utad018 　

ISSN： 1368-4221

eISSN： 1368-423X

More details Close

Abstract We extend the Principal Orthogonal complEment Thresholding (POET) framework by Fan, J., Y. Liao, M. Mincheva (2013) to estimate large covariance matrices with a “mixed” structure of observable and unobservable strong/weak factors, and we call this method the extended POET (ePOET). Especially, the weak factor structure allows the existence of much slowly divergent eigenvalues of the covariance matrix that are frequently observed in real data. Under some mild conditions, we derive the uniform consistency of the proposed estimator for the cases with or without observable factors. Furthermore, several simulation studies show that the ePOET achieves good finite-sample performance regardless of data with strong, weak, or mixed factors structure. Finally, we conduct empirical studies to present the practical usefulness of the ePOET.
Estimating spatial regression models with sample data-points: A Gibbs sampler solution

Giuseppe Arbia, Yasumasa Matsuda, Junyue Wu

Spatial Statistics　47　100568-100568　2022/03
Publisher: Elsevier {BV}
DOI： 10.1016/j.spasta.2021.100568 　
Spatial analysis of subjective well-being in Japan

Anqi Li, Takaki Sato, Yasumasa Matsuda

JAPANESE JOURNAL OF STATISTICS AND DATA SCIENCE　2022/01

DOI： 10.1007/s42081-021-00143-x 　

ISSN： 2520-8756

eISSN： 2520-8764
Discussion on Competition for Spatial Statistics for Large Datasets Invited

Yasumasa Matsuda

JOURNAL OF AGRICULTURAL BIOLOGICAL AND ENVIRONMENTAL STATISTICS　26　(4)　612-613　2021/12

DOI： 10.1007/s13253-021-00463-1 　

ISSN： 1085-7117

eISSN： 1537-2693
A threshold extension of spatial dynamic panel model with fixed effects Peer-reviewed

Junyue Wu, Yasumasa Matsuda

Journal of Spatial Econometrics　2　(1)　2021/12
Publisher: Springer Science and Business Media LLC
DOI： 10.1007/s43071-021-00008-1 　

ISSN： 2662-2998

eISSN： 2662-298X
Log-periodogram regression of two-dimensional intrinsically stationary random fields Peer-reviewed

Yoshihiro Yajima, Yasumasa Matsuda

Japanese Journal of Statistics and Data Science　3　(1)　333-347　2020/06/20
Publisher: Springer Science and Business Media {LLC}
DOI： 10.1007/s42081-020-00078-9 　

ISSN： 2520-8756

eISSN： 2520-8764
Special feature: spatial statistics Invited

Yasumasa Matsuda

Japanese Journal of Statistics and Data Science　3　(1)　103-105　2020/06
Publisher: Springer Science and Business Media LLC
DOI： 10.1007/s42081-020-00081-0 　

ISSN： 2520-8756

eISSN： 2520-8764
Evaluation of regional variations in healthcare utilization Peer-reviewed

Yoko Ibuka, Yasumasa Matsuda, Keishi Shoji, Tsukasa Ishigaki

Japanese Journal of Statistics and Data Science　3　(1)　349-365　2020/06
Publisher: Springer Science and Business Media LLC
DOI： 10.1007/s42081-020-00082-z 　

ISSN： 2520-8756

eISSN： 2520-8764
Locally stationary spatio-temporal processes Peer-reviewed

Yasumasa Matsuda, Yoshihiro Yajima

JAPANESE JOURNAL OF STATISTICS AND DATA SCIENCE　1　(1)　41-57　2018/06

DOI： 10.1007/s42081-018-0003-9 　

ISSN： 2520-8756

eISSN： 2520-8764
Spatial Autoregressive Conditional Heteroskedasticity Models Peer-reviewed

Takaki Sato, Yasumasa Matsuda

JOURNAL OF THE JAPAN STATISTICAL SOCIETY　47　(2)　221-236　2017/12/28
Publisher: The Japan Statistical Society
DOI： 10.14490/jjss.47.221 　

ISSN： 1882-2754

eISSN： 1348-6365
Generalized Nelson-Siegel term structure model: do the second slope and curvature factors improve the in-sample fit and out-of-sample forecasts? Peer-reviewed

Wali Ullah, Yasumasa Matsuda, Yoshihiko Tsukuda

JOURNAL OF APPLIED STATISTICS　42　(4)　876-904　2015/04

DOI： 10.1080/02664763.2014.993363 　

ISSN： 0266-4763

eISSN： 1360-0532
Dynamics of the term structure of interest rates and monetary policy: is monetary policy effective during zero interest rate policy? Peer-reviewed

Wali Ullah, Yasumasa Matsuda, Yoshihiko Tsukuda

JOURNAL OF APPLIED STATISTICS　41　(3)　546-572　2014/03

DOI： 10.1080/02664763.2013.845142 　

ISSN： 0266-4763

eISSN： 1360-0532
Term Structure Modeling and Forecasting of Government Bond Yields: Does a Good In-Sample Fit Imply Reasonable Out-of-Sample Forecasts?

Wali Ullah, Yasumasa Matsuda, Yoshihiko Tsukuda

Economic Papers　32　(4)　535-560　2013/12/01
Publisher: Wiley-Blackwell
DOI： 10.1111/1759-3441.12046 　

ISSN： 1759-3441 0812-0439
Term Structure Forecasting of Government Bond Yields with Latent and Macroeconomic Factors: Do Macroeconomic Factors Imply Better Out-of-Sample Forecasts? Peer-reviewed

Wali Ullah, Yoshihiko Tsukuda, Yasumasa Matsuda

JOURNAL OF FORECASTING　32　(8)　702-723　2013/12

DOI： 10.1002/for.2266 　

ISSN： 0277-6693

eISSN： 1099-131X
Broadband semi-parametric estimation of long-memory time series by fractional exponential models Peer-reviewed

Masaki Narukawa, Yasumasa Matsuda

JOURNAL OF TIME SERIES ANALYSIS　32　(2)　175-193　2011/03

DOI： 10.1111/j.1467-9892.2010.00690.x 　

ISSN： 0143-9782
ON NONPARAMETRIC AND SEMIPARAMETRIC TESTING FOR MULTIVARIATE LINEAR TIME SERIES Peer-reviewed

Yoshihiro Yajima, Yasumasa Matsuda

ANNALS OF STATISTICS　37　(6A)　3529-3554　2009/12

DOI： 10.1214/08-AOS610 　

ISSN： 0090-5364
Selecting models with different spectral density matrix structures by the cross-validated log likelihood criterion Peer-reviewed

Y Matsuda, Y Yajima, H Tong

BERNOULLI　12　(2)　221-249　2006/04

DOI： 10.3150/bj/1145993973 　

ISSN： 1350-7265
On testing for separable correlations of multivariate time series Peer-reviewed

Y Matsuda, Y Yajima

JOURNAL OF TIME SERIES ANALYSIS　25　(4)　501-528　2004/07

DOI： 10.1111/j.1467-9892.2004.01795.x 　

ISSN： 0143-9782
非線形回帰モデルによるヘドニック・アプローチ Invited Peer-reviewed

松田安昌

住宅土地経済　52　(52)　29-35　2004/03
Publisher:
ISSN： 0917-3498
A spatio-temporal regression model for the analysis of functional MRI data. International-journal Peer-reviewed

Kota Katanoda, Yasumasa Matsuda, Morihiro Sugishita

NeuroImage　17　(3)　1415-28　2002/11

DOI： 10.1006/nimg.2002.1209 　

ISSN： 1053-8119
A test of linearity against functional coefficient autoregressive models Peer-reviewed

Yasumasa Matsuda

Communications in Statistics - Theory and Methods　28　(11)　2539-2551　1999/01
Publisher: Informa UK Limited
DOI： 10.1080/03610929908832437 　

ISSN： 0361-0926

eISSN： 1532-415X
A diagnostic statistic for functional-coefficient autoregressive models Peer-reviewed

Yasumasa Matsuda

Communications in Statistics - Theory and Methods　27　(9)　2257-2273　1998/01
Publisher: Informa UK Limited
DOI： 10.1080/03610929808832226 　

ISSN： 0361-0926

eISSN： 1532-415X
A NONPARAMETRIC TEST FOR NONLINEARITY BY THE WEIGHTED LEAST SQUARES METHOD Peer-reviewed

Yasumasa Matsuda

JOURNAL OF THE JAPAN STATISTICAL SOCIETY　27　(2)　141-156　1997
Publisher: The Japan Statistical Society
DOI： 10.14490/jjss1995.27.141 　

ISSN： 1882-2754

eISSN： 1348-6365

More details Close

We propose nonparametric time domain statistics to test time series nonlinearity. We show the asymptotic properties of our statistics for an autoregressive process and also discuss the asymptotic results for some kinds of nonlinear processes. Next, we show the power of our statistics for various time series and compare the power of our statistics with that of other well known nonparametric statistics proposed by McLeod and Li [7], Tsay [11] and Hjellvik and Tjøstheim [12] by simulation studies.

Show all ︎Show first 5

Misc. 26

東京23区賃貸不動産市場の深層学習生存時間分析

佐藤圭悟, 松田安昌

統計関連学会連合大会　2025/09
Density on density regression by spline mixtures

Yasumasa Matsuda, Rei Iwafuchi

Tohoku-NTU Joint Seminar　2025/01
Spatial factor models for surface time series Invited

Yasumasa Matsuda

2024 First Macau International Conference on Business Intelligence and Analytics　2024/12
Fourier Analysis of Surface Time Series Invited

Time Series, Random Fields and beyond　2024/09
Deep learning extensions of Cox regression and their applications to rental property market in Tokyo Invited

Yasumasa Matsuda

10th Annual Data Science in Engineering and Life Sciences Symposium　2024/08
Spatial factor models for surface time series Invited

Spatial factor models, for surface time series

EcoSta 2024　2024/08
Spatial Factor Models Invited

Yasumasa Matsuda, Runyu Dai

York-Tohoku Symposium　2024/06
Estimation of high-dimensional volatility matrices by spatial AR models with unknown linkages

Yasumasa Matsuda, Runyu Dai, Jieying Zhang, Rei Iwafuchi

Spatial Econometrics Association　2024/06
Functional PCA of human population in Tokyo Invited

NTU-Tohoku U 7th Symposium on AI and Human Studies, Taipei　2024/03
Fourier analysis of spatio temporal data Invited

IMS-APRM 2024　2024/01
高次元ファイナンス時系列におけるボラティリティ推定

松田安昌, 岩淵澪

統計関連学会連合大会　2023/09
Deep learning for multivariate volatility forecasting in high-dimensional financial time series

Yasumasa Matsuda, Rei Iwafuchi

2023/08
Deep learning for multivariate volatility forecasting in high-dimensional financial time series Invited

Yasumasa Matsuda, Rei Iwafuchi

2023/08
How long do voluntary lockdowns keep people at home? The role of social capital during the COVID-19 pandemic.

Yuta Kuroda, Takaki Sato, Yasumasa Matsuda

Data Science and Service Research Discussion Paper　(125)　2022/05
Levy Driven CARMA Processes: Properties, Applications and Inference

松田安昌

日本統計学会誌　49　(2)　2020

ISSN： 0389-5602
WAVELET ANALYSIS OF IRREGULARLY SPACED DATA AND ITS SPATIO-TEMPORAL EXTENSION

Matsuda Yasumasa

DSSR Discussion Papers　(30)　1-21　2014/11
Publisher: 東北大学大学院経済学研究科
Generalized Whittle Estimate for Nonstationary Spatial Data

Matsuda Yasumasa

DSSR Discussion Papers　(9)　1-20　2013/05
Publisher: 東北大学大学院経済学研究科
時空間統計解析:新たなる分野横断的展開一般化Whittle法による不等間隔時空間データの分析

松田安昌

統計数理　60　(1)　2012

ISSN： 0912-6112
Extended Local Whittle Estimation of a Long-memory Stochastic Volatility Model

生川雅紀, 松田安昌

日本統計学会誌　39　(2)　2010

ISSN： 0389-5602
Spectral analysis of irregularly spaced data on R[d]

Annual report of the Economic Society,Tohoku University　68　(3)　483-496　2007/03
Publisher: 東北大学経済学会
ISSN： 0387-3056
Fourier Analysis of Irregularly Spaced Data on Rd

Yasumasa Matsuda, Yoshihiro Yajima

Japanese journal of applied statistics　36　(1)　1-14　2007
Publisher: Japanese Society of Applied Statistics
DOI： 10.5023/jappstat.36.1 　

ISSN： 0285-0370

eISSN： 1883-8081
Analysis of Price of Secondhand Apartments in Sendai City by Spatial Band Spectrum Regression

久保田茂裕, 松田安昌

日本統計学会誌　37　(1)　2007

ISSN： 0389-5602
C-1 多変量時系列におけるノンパラメトリックおよびセミパラメトリック検定について(経済時系列)(2003年度統計関連学会連合大会記録(日本統計学会第71回大会))

矢島美寛, 松田安昌

日本統計学会誌　33　(3)　384-384　2003
Publisher: 日本統計学会
ISSN： 0389-5602
On Testing for Separable Correlations for Multivariate Time Series

Matsuda, Y, Yajima, Y

J.Time Series Analysis　25　501-528　2001

DOI： 10.1111/j.1467-9892.2004.01795.x 　
Articles Statistical Analysis of Some Marketing Data via Dynamic Linear Regression Models

Matsuda Yasumasa

The journal of economics, Niigata University　69　13-21　2000/09
Publisher: Niigata University
ISSN： 0286-1569
ONPARAMETRIC TEST FOR NONLINEARITY BY THE WEIGHTED LEAST SQUARES METHOD

MATSUDA Yasumasa

Journal of the Japan Statistical Society　27　(2)　141-156　1997/12/01
Publisher: 日本統計学会
ISSN： 0389-5602

Show all ︎Show first 5

Presentations 6

Dynamic factor models for density-valued time series Invited

Yasumasa Matsuda

2nd Tohoku NTU joint seminar　2026/03/26
混合治癒モデルを用いたオンラインフリーマーケットの生存時間解析 Invited

込山湧士, 松田安昌, 石原昌和

DEIM2026 第18回データ工学と情報マネジメントに関するフォーラム　2026/03/02
Dynamic factor models for density-valued time series Invited

Yasumasa Matsuda

2026/02/19
Density-Valued ARMA Models by Spline Mixtures Invited

Rei Iwafuchi, Yasumasa Matsuda

2025 IMS International Conference on Statistics and Data Science　2025/12/15
オンラインフリーマーケット市場への生存時間分析の応用

込山湧士, 松田安昌, 石原昌和

NII-IDRユーザフォーラム2025　2025/11/26
空間上の確率密度関数のB-splineによるSieve 推定とその応用

岩淵澪, 松田安昌

2025年度統計関連学会連合大会　2025/09/10

Show all Show first 5

Research Projects 19

時空間系列分析の理論と応用 Competitive

System: Cooperative Research

2005/10 - Present
再生可能エネルギー需給予測にむけた大規模気象予測モデルの構築と社会経済分析

松田安昌

Offer Organization: 日本学術振興会

System: 科学研究費助成事業

Category: 基盤研究(A)

Institution: 東北大学

2026/04/01 - 2030/03/31
Development of standard setting technology which guarantee bone forming ability of jaw bone derived skeltal stem cells

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research

Category: Grant-in-Aid for Scientific Research (B)

Institution: Tohoku University

2025/04/01 - 2028/03/31
spatio-temporal data modeling with applications to social science

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research

Category: Grant-in-Aid for Scientific Research (B)

Institution: Tohoku University

2021/04/01 - 2026/03/31
Modeling of large scale spatio-temporal data and its empirical applications in social science

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research

Category: Grant-in-Aid for Scientific Research (B)

Institution: Tohoku University

2021/04/01 - 2026/03/31
Innovative Developments of Theories and Methodologies for Large Complex Data

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (A)

Category: Grant-in-Aid for Scientific Research (A)

Institution: University of Tsukuba

2020/04/01 - 2025/03/31
空間計量経済学における最重要課題への挑戦と新たな展開

堤盛人, 松田安昌, 瀬谷創, 村上大輔

Offer Organization: 日本学術振興会

System: 科学研究費助成事業基盤研究(A)

Category: 基盤研究(A)

Institution: 筑波大学

2018/04/01 - 2023/03/31

More details Close

空間計量経済学の最重要課題を解決するとともにさらなる発展を目的として、［Ⅰ］空間重み行列の内生化に関する実証研究と新たなモデル開発［Ⅱ］ビッグデータへの新たな対処法の開発［Ⅲ］組成データ解析を応用した空間計量経済学の新たな展開［Ⅳ］研究成果の公開・利用促進と日本における空間計量経済学分野の研究者の育成の四つの課題に取り組んだ。具体的には、［Ⅰ］空間計量経済学における空間重み行列Wの内生性をモデル化する第一段階として、サンプルの偏りを扱うHeckmanモデルと空間計量経済モデルの統合を試みた。また、地理学の分野で提案されたEigenvector Spatial Filtering (ESF)を応用し、新たな空間分位点回帰モデル及び空間可変パラメータモデルを開発した。［Ⅱ］空間計量経済学におけるビッグデータへの対処法として、次元縮減アプローチに着目して、標本数と(ハイパー)パラメータ数の両方に関して線形時間で空間回帰モデルを推定するアルゴリズムを新規開発した。また、そのアルゴリズムを空間計量経済モデルの推定に応用してその精度と計算効率を確認した。さらに、複数の地域において調査された個人レベルの調査データに対して回帰モデルを考える際、回帰係数を全地域で共通とせず地域別にことなるが近い地域では似た値をとるモデルを構築し、推定法を考案した。［Ⅲ］Conditional Autoregressive (CAR) Modelに基づく組成データモデルの開発を進め、空間計量経済学の中で提案されてきた他の様々なモデルをCoDaへ拡張するための方法論の検討を行った。［Ⅳ］空間計量経済学に関するセミナー開催のための企画について検討を行った。また、空間計量経済学に関する国際学会である Spatial Econometrics Association と連携しながら、学会開催の準備を進めた。
Building new models for spatio-temporal econometrics and thier statistical inference

Yajima Yoshihiro

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)

Category: Grant-in-Aid for Scientific Research (B)

Institution: Tohoku University

2017/04/01 - 2022/03/31

More details Close

In this research, in order to contribute the development of spatio-temporal econometirics, we have constructed new spatio-temporal econometric models, clarified their statistical properties and considered their applicability to analyze empirical spatio-temporal data. First for nonstationary spatio-temporal random fields, we constructed new models and derived their statistical properties. Furthemore we showed their usefulness for analyzing empirical data. Secondly we constructed new models for large spatio-temporal data sets and efficient procedures to analyze them. Thirdly for financial data analysis, we proposed new models that take their spatio-temporal dependency into account of and more robust estimators based on rank statistics than the least squares estimator. Finally for uraban and regional economic analysis, especially network society anlysis, we constructed effective models and applied them to analyze an effect of a regulation for land use on land prices.
CARMA random fields and their applications to large spatio-temporal data

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)

Category: Grant-in-Aid for Scientific Research (B)

Institution: Tohoku University

2017/04/01 - 2021/03/31
Statistical modeling of spatio-temporal data on natural and social phenomena, and those understanding

Nishii Ryuei, Koda Satoru

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)

Category: Grant-in-Aid for Scientific Research (B)

Institution: Kyushu University

2015/04/01 - 2019/03/31

More details Close

The following studies were conducted. 1) In IGARSS 2015, a special session about the social application of the remote sensing imagery was orgenized with Columbia University. 2) Disturbance strom index was accurately predicted by regression model based on physical measurements about the solar activity. 3) Gene regulatory network was estated by sparse modeling of regression model on time-series of global gene expressions. 4) For super high-resolution land covering imagery, multi-label classification was proposed by machine learning by the use of hierarchical structure of the output layer and spatial dependency of the input layer.
Modelling of non-stationary spatio-temporal data with estimation, testing and forecasting

Matsuda Yasumasa

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (B)

Category: Grant-in-Aid for Scientific Research (B)

Institution: Tohoku University

2013/04/01 - 2017/03/31

More details Close

We have developed a model to analyze huge spatio-temporal data set with non-stationary structures and proposed a method of parameter estimation, testing of correlation structures and forecasting with applications to empirical data mainly from environmental studies. Specifically, We have considered a separable model whose spatio-temporal correlations are given by product of spatial and temporal correlations. In the empirical studies in this research, we employed continuous autoregressive moving average (CARMA) models for spatial behaviors and traditional ARMA models for temporal ones. We applied the separable models to the US rain fall data recorded monthly at 6000 points irregularly spaced inside US continent and found proposed an efficient estimation, testing and forecasting methods that can cope with huge data set composed of several million data points.
金融緩和政策がおよぼす日本国債市場へのインパクトのモデル分析

松田安昌, ULLAH Wali

Offer Organization: 日本学術振興会

System: 科学研究費助成事業特別研究員奨励費

Category: 特別研究員奨励費

Institution: 東北大学

2013/04/01 - 2015/03/31

More details Close

Nelson-Siegelモデルは金利期間構造のモデル化に伝統的に広く使われてきた。実際、欧米諸国の国債金利の期間構造はNelson-Siegelモデルによって高い精度で説明されることが多くの実証研究によって知られている。一方、失われた２０年とよばれる長い不況からいまだに脱し切れていない日本経済において、日本国債の金利は非常に特殊な特徴を持つ。日本銀行のゼロ金利政策により短期金利はほぼゼロであるが、長期金利は複数の変曲点をもつ。Nelson-Siegelモデルはこれらの日本国債金利の特殊な変動を説明することができない。そこで本研究では、Nelson-Siegelモデルを一般化して２つのスロープパラメータを時変構造にし、さらに誤差項のボラティリティをEGARCHモデルによって分散の不均一構造を持たせた。この一般化Nelson-Siegelモデルの良さを検証するために、1)データへのあてはまりの良さ、2)予測量とその予測誤差、の２点を計算して既存のNelson-Siegelモデルの結果と比較を行った。その結果、提案した一般化Nelson-Siegelモデルは日本国債のイールドカーブへのあてはまりを改善するのみならず、国債金利の予測パフォーマンスを様々な満期設定の下で改善することがわかった。さらに、外部の経済ショックに対し、プラスのショックとマイナスのショックはボラティリティに対称でない影響を与え、過去のショックより直近の経済ショックが現在のボラティリティに影響していることを示した。
Theory of spatio-temporal data analysis and its applications to environmental data

MATSUDA Yasumasa

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (C)

Category: Grant-in-Aid for Scientific Research (C)

Institution: Tohoku University

2009/04/01 - 2014/03/31

More details Close

This research has conducted a proposition of nonstationary Fourier analysis of spatio-temporal data when observation points are irregularly spaced. Originally nonstatinary Fourier analysis was proposed for nonstationary time series analysis by local periodogram in order to identify temporal dependencies of time series models. Here we extend the method for time series to that for spatio-temporal data that makes it possible to identify local dependencies of spatial model parameters. The consistency of the estimators by the nonstationary Fourier methods has been proved. We have applied the methodology to land price data in Kanto area and examined nonstationary behaviors of land price data by the Fourier analysis.
On a comprehensive research on statistical modelling and analysis for spatio-temporal data

YAJIMA Yoshihiro, MASE Shigeru, SHIMIZU Kunio, NISHII Ryuei, KISHINO Hirohisa, KURIHARA Kouji, YOSHIDA Atsushi, KUBOKAWA Tatsuya, FUKUCHI Jun=ichiro, NAGAI Keiji, NISHIYAMA Yoshihiko, MATSUDA Yasumasa, MARUYAMA Yuzo, TAKAHASHI Kunihiko, OGATA Yoshihiko

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (A)

Category: Grant-in-Aid for Scientific Research (A)

Institution: The University of Tokyo

2007 - 2010

More details Close

In an interdisciplinary way based on statistical science methodology, we constructed models for spatio-temporal data, which interact with each other spatially or spatio-temporally, discussed both their theoretical properties and applicability for empirical data
時系列相関が存在する場合の統計的因果性分析法について

松田安昌

Offer Organization: 日本学術振興会

System: 科学研究費助成事業若手研究(B)

Category: 若手研究(B)

2004 - 2006

More details Close

時系列グラフィカルモデリングについて平成17年度までに出した成果を2本の論文にまとめ、Bernoulli, Biometrikaの2誌に発表した。Bernoulliでは、モデル選択基準: Cross Validateed log likelihood criterion (CVLL 交差確認対数尤度基準)を用いたグラフィカルモデリング法を提案し、Biometrikaでは検定統計量を用いた同定法を提案した。統計モデリングでは、モデル選択を使うか、検定統計量を使うか、の二通りの方法があるが、時系列グラフィカルモデリングにおいても、二種の同定法を提案したことになる。二種の方法は互いに補完し合う性質を持っていて、どちらかが常に他方を優越することはなく、現実的には両方法を試みてモデル同定を行っていく。次に、上記論文で発表したグラフィカルモデリング同定法を実行するプログラムパッケージをC言語で作成し、公開した。本パッケージでは定常化した多変量時系列を入力すれば、先の述べた二種の同定法によって二通りのグラフィカルモデルが出力される。実行上、注意点がいくつかある。まず多変量時系列を定常化する必要がある。本方法では多変量時系列に定常性を仮定しているので、そのままの形で入力しても有意なグラフィカルモデルを同定できない。通常は、差分をとるか移動平均を用いて定常かを行う。次に、両方法とも手続き中にスペクトル密度関数のスムージングを行う箇所があるが、必ずしも最適なものではない。最後に、検定統計量による方法では。モデリングを行う前に有意水準を決定しておかなければならない。通常は5%に設定するが、必ずしも根拠のある数字でない。
Statistical Analysis of Spatio-Temporal Data: its Theory and Applications

YAJIMA Yoshihiro, OGATA Yoshihiko, KUBOKAWA Tatsuya, NISHII Ryuei, MATSUDA Yasumasa, MARUYAMA Yuzo

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research Grant-in-Aid for Scientific Research (A)

Category: Grant-in-Aid for Scientific Research (A)

Institution: The University of Tokyo

2003 - 2006

More details Close

We held a research conference on spatio-temporal statistical analysis every year through the research period from 2003 to 2006. These conferences have made much contribution to the development of spatio-temporal statistical analysis in both theory and its applications and to that the researchers in this field are able to deepen the understanding of the topic and share the studies to be solved in future with each other. We have conducted our research project by dividing the members into five groups. First for inference theory and applications of random fields, we proposed a generalization of Fourier analysis of time series to spatio-temporal data and applied it to analyze a land price data of Tokyo Metropolitan area. Secondly for inference theory and its applications of spatio-temporal point processes, we developed an intensity function taking into account of spatio and temporal correlations simultaneously and applied it to analyze earthquake data. Furthermore we developed a threshold method to predict the sample variance and applied it to analyze rainfall data. Spatio-temporal correlation analysis between epidemiological data was also conducted. Thirdly for inference theory and applications of large sample-size data, we developed a new methodology of dscriminat analysis on Markov random fields and Echelon method for detecting hot spots of spatial data. Fourthly for inference theory and its application of small area data, we developed minimax and empirical Bayes estimators, which are able to overcome multicolinearity, and are robust against the instability caused by small sample size data. Finally for inference theory and its applications of panel data, we conducted an empirical analysis of spatial competition and its equiburium. And we conducted simulation studies, which investigated spurious spatial correlation caused by autocorrelations within individual time series and spatial heteroskedasticity between individual time series.
時空間モデルの理論と環境データ等への応用

松田安昌, 矢島美寛

Offer Organization: 日本学術振興会

System: 科学研究費助成事業基盤研究(C)

Category: 基盤研究(C)

Institution: 新潟大学

2002 - 2004

More details Close

時空間データ分析の理論として、時空間データ相関構造の検定法を提案し、実データ等への応用として、fMRIデータ解析および地価データの分析を行った。時空間データの相関構造の検定とは、2種のスペクトルを考え、どちらのスペクトルがデータにふさわしいかを定めることである。今年度の研究では、2種のスペクトルとして多くの応用可能性があるクラスを考え、択一を行う統計量をカルバック・ライブラーの距離を導入することで考案した。この検定統計量により適用可能な例として、複数の系列間の因果性の分析がある。さらに、統計量の漸近分布を導出し、検定手順を確立した。実データヘの応用では、まずfMRIデータヘ時空間データ解析法を適用した新しい分析法を提案した。そこでは誤差が時空間モデルにしたがう回帰モデルを提案し、誤差の相関構造をノンパラメトリックに扱って回帰係数を推定する方法を使った。この方法を実際のfMRIデータに適用してみた結果、有効な方法であることが確かめられた。最後に地価データ分析を行った。関東一円の16851地点の2001年度公示地価データを収集し、商業施設、公園・緑地の地価への効果を回帰分析により測定した。標本地点間の相関関係をどのように表現するかが課題であるが、ここではfunctional-coefficient regression modelという非線形回帰モデルを適用し、相関関係を表現した。この非線形回帰分析により、土地利用面積の地価への効果と標本地点からの距離の関係を分析した。
Research on Reverse Logistics Including Recycle and Reuse

SARASAWA Yutaka, MATSUDA Yasumasa, AIURA Nobunori, WAKABAYASHI Kenzo, UCHIDA Satoshi, SATO Katsunao

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research

Category: Grant-in-Aid for Scientific Research (C)

Institution: Kanagawa Univ.

1999 - 2002

More details Close

We have published 9 academic papers, including 8 papers in abroad and one paper in domestic as find results. In addition, total of 24 papers of academic proceedings, have been made presentation of the academic conference and one logistics look entitled with "Logistics and Environments". But of 24 papers of the proceedings, mainly of the Society of Logistics Engineers in U.S.A International Society of Logistics Engineers in U.S.A and Europe and The Japan, Society of Logistics Systems, 8 papers have been published in U.S.A and Europe and other 16 papers and have been published in Japan. As to reports for this research of The Ministry of Education, Culture, Sports, Science and Technology (MEXT), we have produced Three reports; Report of Summary Data in 2001 (Total Page: 182), Interview Report in 2002 (Total Page: 643) and Final Report in 2003. All the reports include not only research results of reverse logistics with recycle, reuse and disposal research but also other logistics managements concerned. In this sense research results have brought about from view points of contribution to wider ranges of logistics management as well as reverse logistics. As final achievements of this research the following items are summarized; ・ have contributed to enhancement of research activities in the field of reverse logistics ・ have developed a new frontier of logistics management as well as reverse logistics ・ have nourished research activities of younger research in this fields, concerned ・ have disseminated research results to academicians and industrial expertise in the world
A Research on A Development Model of Corporate Logistics Concepts and Purposes

KARASAWA Yutaka, MATSUDA Yasumasa, UCHIDA Satoshi, WAKABAYASHI Keizo, KITAOKA Masatoshi

Offer Organization: Japan Society for the Promotion of Science

System: Grants-in-Aid for Scientific Research

Category: Grant-in-Aid for Scientific Research (C)

Institution: Kanagawa University

1996 - 1998

More details Close

Research on corporate logistics concepts and purposes has been paid little attention to by both academicians and researchers in the fields of logistics. Therefore this study is of great importance in terms of a brand new research in the field of logistics and we are only one group who has made research twice in the past and in that sense questionnaires have been mailed out to industrial firms, 700 members of Japan Institute of Logistics Systems concerned in 1991 and 1996, respectively. The main purposes are to propose a hypothetical model for development process of corporate logistics concepts and purposes based on analysis of survey results, to investigate if the proposed hypothetical model is correct or not based on the statistical figures, to estimate present and future development stage of corporate logistics concepts and purposes in Japan based on questionnaires to make time comparison in terms of differences of results between survey of 1991 and that of 1995 and to make clear historical and future trends on corporate logistics concepts and purposes for three different periods, 1900, 1995 and 2000. As a result, our proposed conceptual model is justified and development level of Japan phases are confirmed and time sequence comparison of an analysis is made.

Show all Show first 5