Updated on 2024/03/30

写真a

 
KATAYAMA,Naoya
 
Organization
Faculty of Economics Professor
Title
Professor
Contact information
メールアドレス
External link

Degree

  • Essays on Seasonally and Fractionally Differenced Time Series ( 2004.3 )

Research Areas

  • Humanities & Social Sciences / Economic statistics

Research History

  • Department of Economics, Kyushu Univ.

    2007.4 - 2010.3

      More details

Professional Memberships

Papers

  • The portmanteau tests and the LM test for ARMA models with uncorrelated errors Reviewed

    Naoya Katayama

    Advances in Time Series Methods and Applications: the A. Ian McLeod Festschrift. Editors: W. K. Li, David Stanford and Hao Yu, Springer   131-150   2016

     More details

  • Proposal of Robust M Tests and Their Applications

    Naoya Katayama

    Woking Paper Series, Economic Society of Kansai University   F-65   2013

     More details

  • Chi-squared portmanteau tests for structural VARMA models with uncorrelated errors

    Naoya Katayama

    JOURNAL OF TIME SERIES ANALYSIS   33 ( 6 )   863 - 872   2012.11

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:WILEY-BLACKWELL  

    We consider portmanteau tests for testing the adequacy of structural vector autoregressive moving average models with uncorrelated errors. Under the assumption that errors are uncorrelated but non-independent, it is known that the LjungBox (or BoxPierce) portmanteau test statistic is asymptotically distributed as a weighted sum of chi-squared random variables which can be far from the chi-square distribution usually employed. We therefore propose a new portmanteau statistic that is asymptotically chi-squared even in the presence of uncorrelated but non-independent errors. Monte Carlo experiments illustrate the finite sample performance for the proposed portmanteau test.

    DOI: 10.1111/j.1467-9892.2012.00799.x

    Web of Science

    researchmap

  • Robust M Tests Using Projection Matrices

    Naoya Katayama

    Woking Paper Series, Economic Society of Kansai University   F-56   2012

     More details

  • Chi-Squared Portmanteau Tests for Weak Vector Autoregressive Models

    Naoya Katayama

    Woking Paper Series, Economic Society of Kansai University   F-54   2012

     More details

  • Chi-Squared Portmanteau Statistics for Vector Autoregressive Models with Uncorrelated Errors

    Naoya Katayama

    Kansai University Review of Economics   No 13, 1-23   2011

     More details

  • Chi-Squared Portmanteau Tests for Weak Vector Autoregressive Models

    Naoya Katayama

    Woking Paper Series, Economic Society of Kansai University   F-47   2011

     More details

  • On Multiple Portmanteau Tests, Reviewed

    Naoya Katayama

    Journal of Time Series Analysis   2009

     More details

  • 合理的バブルの検定の検出力について

    片山直也

    Discussion Paper Series, Faculty of Economics, Kyushu University   2009-4   2009

     More details

    revised: Woking Paper Series J-27, Economic Society of Kansai University, 2010.

    researchmap

  • Simulation Studies of Multiple Portmanteau Tests

    Naoya Katayama

    Discussion Paper Series, Faculty of Economics, Kyushu University   2009-4   2009

     More details

  • Asymptotic Prediction of Mean Squared Error for Long-Memory Processes, with Estimated Parameters

    Naoya Katayama

    JOURNAL OF FORECASTING   27 ( 8 )   690 - 720   2008.12

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:JOHN WILEY & SONS LTD  

    In this paper we deal with the prediction theory of long-memory time series. The purpose is to derive a general theory of the convergence of moments of the nonlinear least squares estimator so as to evaluate the asymptotic prediction mean squared error (PMSE). The asymptotic PMSE of two predictors is evaluated. The first is defined by the estimator of the differencing parameter, while the second is defined by a fixed differencing parameter: in other words, a parametric predictor of the seasonal autoregressive integrated moving average model. The effects of misspecifying the differencing parameter is a long-memory model are clarified by the asymptotic results relating to the PMSE. The finite sample behaviour of the predictor and the model selection in terms of PMSE of the two predictors are examined using simulation, and the source of any differences in behaviour made clear in terms of asymptotic theory. Copyright (c) 2008 John Wiley & Sons, Ltd.

    DOI: 10.1002/for.1078

    Web of Science

    researchmap

  • An improvement of the portmanteau statistic

    Naoya Katayama

    JOURNAL OF TIME SERIES ANALYSIS   29 ( 2 )   359 - 370   2008.3

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:BLACKWELL PUBLISHING  

    The portmanteau statistic is based on the first m-residual autocorrelations, and is used for diagnostic checks on the adequacy of fit of a model. In this article, we propose a modified portmanteau statistic with a correction term that allows for the use of small values of m for the chi-squared test. For this modification, we take a different approach to that suggested by Ljung [Biometrika (1986), Vol. 73, pp. 725-30]. Their empirical behaviour is clarified using asymptotic theory.

    DOI: 10.1111/j.1467-9892.2007.00559.x

    Web of Science

    researchmap

  • Portmanteau Likelihood Ratio Tests for Morel Selection

    Naoya Katayama

    Discussion Paper Series, Faculty of Economics, Kyushu University   2008-1   2008

     More details

  • On Multiple Portmanteau Tests

    Naoya Katayama

    Discussion Paper Series, Faculty of Economics, Kyushu University   2008-4   2008

     More details

  • Seasonally and fractionally differenced time series

    Naoya Katayama

    HITOTSUBASHI JOURNAL OF ECONOMICS   48 ( 1 )   25 - 55   2007.6

     More details

    Language:English   Publishing type:Research paper (scientific journal)   Publisher:HITOTSUBASHI ACAD  

    This paper presents a generalized seasonally integrated autoregressive moving average (SARIMA) model that allows the two differencing parameters to take on fractional values. We examine the asymptotic properties of the estimators and test statistics when the mean of the model is unknown. The findings show that standard asymptotic results hold for the tests and that the conditional sum of squares estimators are consistent and tends towards normality. The paper provides a modelling application using data on total power consumption in Japan.

    Web of Science

    researchmap

  • 予測の平均二乗誤差を基準とするモデル選択について Reviewed

    片山直也

    統計数理   第54 巻第2 号,pp.481-510   2006

     More details

  • Essays on Seasonally and Fractionally Differenced Time Series

    Naoya Katayama

    2004

     More details

  • Seasonally and Fractionally Differenced Time Series

    Naoya Katayama

    ディスカッションペーパー, 一橋大学21世紀COEプログラム社会科学の統計分析拠点   No.11   2004

     More details

    Published in Hitotsubashi Journal of Economics, Vol.48, No.1, 25-55

    researchmap

  • Asymptotic Prediction Mean Squared Error for Strongly Dependent Processes with Estimated Parameters

    Naoya Katayama

    ディスカッションペーパー, 一橋大学21世紀COEプログラム社会科学の統計分析拠点   No.10   2004

     More details

    Published in Journal of Forecasting, 27(8), 690-720

    researchmap

▼display all

Books

  • かばん検定の新展開

    片山直也( Role: Sole author)

    前川功一,得津康義 編著 「金融時系列分析の理論と応用」 広島経済大学研究双書 第39冊  2012 

     More details

  • 実例とEXCELによる統計学トレーニング

    片山直也( Role: Sole author)

    牧野書店  2009 

     More details

Presentations

  • Comments on A Top-Down Method for Rational Bubbles: Application of the Threshold Bounds Testing Approach

    Naoya Katayama

    日本ファイナンス学会第24回大会  2016.5 

     More details

  • Identification and Goodness of Fit Tests for SVAR Models with Application to the Effects of the Quantitative Easing Policy by the Bank of Japan

    Naoya Katayama

    XXIV International Rome Conference on Money, Banking and Finance  2015.12 

     More details

  • Comments on Insight from a Bayesian VAR model with drifting parameters of the French housing and credit markets

    Naoya Katayama

    XXIV International Rome Conference on Money, Banking and Finance  2015.12 

     More details

  • Proposal of Two Robust M tests

    Naoya Katayama

    2012 (EC)2 Conference  2012.12 

     More details

  • Chi-Squared Portmanteau Statistics for Vector Autoregressive Models with Uncorrelated Errors

    Naoya Katayama

    Hitotsubashi Conference on Econometrics 2010  2010.11 

     More details

  • 無相関の誤差項を持つVARモデルのかばん検定統計量の改良

    片山直也

    統計関連学会 連合大会(日本統計学会 2010年度第78回大会)  2010.9 

     More details

  • Modeling Explosive Autoregressive Time Series Under Weak Dependence

    Naoya Katayama

    関西計量経済学研究会2009年度研究発表会  2010.1 

     More details

  • On Multiple Portmanteau Tests

    Naoya Katayama

    International Conference on Econometrics and the World Economy  2009.3 

     More details

  • On Multiple Portmanteau Tests

    Naoya Katayama

    関西計量経済学研究会2008年度研究発表会  2009.1 

     More details

  • On Multiple Portmanteau Tests

    Naoya Katayama

    統計関連学会 連合大会(日本統計学会 2008年度第76回大会)  2008.9 

     More details

  • Portmanteau Likelihood Ratio Tests for Morel Selection

    Naoya Katayama

    Far Eastern and South Asian Meeting of the Econometric Society  2008.7 

     More details

  • モデル選択のための簡単な尤度比検定の提案

    片山直也

    関西計量経済学研究会2007年度研究発表会  2008.2 

     More details

  • 長期記憶時系列の金融・経済分野への応用

    片山直也

    第10回 情報論的学習理論ワークショップ (IBIS 2007)  2007.11 

     More details

  • かばん検定統計量のバイアスについて

    片山直也

    統計関連学会 連合大会(日本統計学会 2007年度第75回大会)  2007.9 

     More details

  • Comments on Likelihood Analysis of Weak Exogeneity in I(2) Systems and reduced Econometric Representations by Takamitsu Kurita

    Naoya Katayama

    日本経済学会2007年度秋季大会  2007.9 

     More details

  • On the Bias of the Portmanteau Statistic

    Naoya Katayama

    The Third Symposium on Econometric Theory and Applications (SETA2007)  2007.4 

     More details

  • 予測の平均二乗誤差を基準とするモデル選択法の考察とAICの解釈

    片山直也

    統計関連学会 連合大会(日本統計学会 2006年度第74回大会)  2006.9 

     More details

  • 平均未知の季節性と長期性を持つ時系列の推定と検定

    片山直也

    統計関連学会 連合大会(日本統計学会 2005年度第73回大会)  2005.9 

     More details

  • 予測の観点からみた「長期記憶モデル vs (S)ARIMAモデル」

    片山直也

    統計関連学会 連合大会(日本統計学会 2004年度第72回大会)  2004.9 

     More details

  • 長期記憶時系列分析について

    片山直也

    一橋大学21世紀COEプログラム社会科学の統計分析拠点の構築のレクチャーシリーズ,2003年度第1回  2004.2 

     More details

  • Seasonal and Fractional Differencing Time Series

    Naoya Katayama

    統計関連学会 連合大会(日本統計学会 2002年度第70回大会)  2002.9 

     More details

▼display all

Research Projects

  • 時系列解析による合理的バブルの発生と崩壊現象の推測、並びに波及経路の解明

    Grant number:20K01600  2020.4 - 2025.3

    日本学術振興会  科学研究費助成事業  基盤研究(C)

    片山 直也

      More details

    Grant amount:\3900000 ( Direct Cost: \3000000 、 Indirect Cost:\900000 )

    昨年から開始した研究「バブルの発生や崩壊が含まれるときの共和分回帰モデルの推定問題」を進めた。この研究は、感染症の分野のSIRモデルとも関係している。このモデルでは、少なくとも感染の初期では、陽性者数の指数級数的な増大に連動して、重症患者数が増大するモデルとなり、これは2つの指数級数的な増大する確率過程が、共和分となっている現象と解釈できそうである。また、共和分関係も、フェーズとともに変化すると考えられる。
    今年度は、共和分ベクトルとラグの最小二乗推定量の一致性を示した。さらに拡張についてもいくつか考えた。例えば、複数のバブルがあった場合への拡張や、KPSS検定への応用、共和分ベクトルがフェーズとともに変化した場合の拡張である。この最後の拡張のモデルは、SIRモデルとも解釈できるが、経済理論でもS&P500のバブル崩壊が、石油価格高騰につながったとする、Caballero, Farhi, and Gourinchas (2008, Brookings Papers on Economic Activity)にも通じるモデルで非常に興味深い。
    すでにシミュレーションも行い、妥当な結果が出たため、理論結果は正しいと思われ、ワーキングペーパーにするため、執筆中である。これら研究の一部は、オンラインではあるが、2021年1月には、関西計量経済学会研究会でも講演を行った。コロナ禍で研究費を使う機会が少ないが、今後は、論文の校閲や投稿費、学会参加費などで使用し、いまの研究をより、精密に仕上げるとともに、実証研究ができないか検討しようと思う。

    researchmap

  • Construction and application of robust M test under non-regularity conditions

    Grant number:26380278  2014.4 - 2019.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    Katayama Naoya

      More details

    Grant amount:\4550000 ( Direct Cost: \3500000 、 Indirect Cost:\1050000 )

    (1) I proposed a robust portmanteau test for a linear time series model in the sense of the assumption of the error term by KVB approach. (2) I examined the recent easy economy policy by bank of Japan by using SVAR models. (3) I found that the rational bubble model yield a (explosive) unit root model with additive outliers (4) Based on (3) I investigated whether unit root tests is robust to additive outliers.

    researchmap

  • Monitaring of parameter chamge in economic time series model

    Grant number:26380279  2014.4 - 2017.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (C)

    MAEKAWA Koichi, LEE Sangyeol, Kusdhianto Setiawan, Amirullah Setya Hardi, Alessio Moneta

      More details

    Grant amount:\4680000 ( Direct Cost: \3600000 、 Indirect Cost:\1080000 )

    The Main results of our study are as follows : (1)Causal inference of economic variables is studied through the independent component analysis(ICA). As a result we could assign an appropriate order of variables in the structural vector autoregression (SVAR) model. This can be used to evaluate economic policy such as quantitative easing policy by Bank of Japan. (2) Statistical methods of analyzing high frequency time series data is developed. Those methods are useful for risk analysis of financial investment. (3) Financial bubble model is developed. It seems to mimic the real bubble phenomenon in Japanese Bubble period very well.

    researchmap

  • Statistical inference for extended models in financial time series

    Grant number:23330075  2011.4 - 2014.3

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Scientific Research (B)

    MAEKAWA Koichi, TOKUTSU Yasuyoshi, KAWAI Ken-ichi, MORIMOTO Tkayuki, KATAYAMA Naoya, NAGATA Shuichi

      More details

    Grant amount:\9880000 ( Direct Cost: \7600000 、 Indirect Cost:\2280000 )

    We studied problems in statistical analysis of financial time series. Since those problems cannot be dealt with classical statistical theory and methods a new research field in econometrics has emerged. As the results the appropriate theory and methods have been developed for problems concerning to the keyword listed below. But even now we think there remain many unsolved problems for compound problems related to plural keywords, such as estimation problems related to vector error correction model with GARCH error, long memory in GARCH process, structural change in high frequency data, modeling of realized volatility by using high frequency time series and so on. We challenged to such problems and attained some significant results including a proposal of some suitable method to such new problems. In addition we evaluated our theoretical and methodological results by computer simulation and applied them to real data.

    researchmap

  • Verification of the rational bubbles by time series analysis

    Grant number:21730175  2009 - 2012

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Young Scientists (B)

    KATAYAMA Naoya

      More details

    Grant amount:\3250000 ( Direct Cost: \2500000 、 Indirect Cost:\750000 )

    To apply to modern financial data, the minister of the research tried to develop the method for M tests which includes the portmanteau tests and is robust not only to heteroscedasticity and serial correlations of unknown form but also to the presence of an estimation effect. The test is a robust M test using the method of Kiefer, Vogelsang, and Bunzel (2000, Econometrica) (referred to as KVB). As applications, we consider robust LM tests, portmanteau tests, GMM over-identification tests and the Hausman tests.

    researchmap

  • Testing for the rational bubble in time series analysis

    Grant number:19830044  2007 - 2008

    Japan Society for the Promotion of Science  Grants-in-Aid for Scientific Research  Grant-in-Aid for Young Scientists (Start-up)

    KATAYAMA Naoya

      More details

    Grant amount:\2973000 ( Direct Cost: \2580000 、 Indirect Cost:\393000 )

    researchmap

▼display all

Devising educational methods

  • 2009年発行の拙著「実例とEXCELによる統計学トレーニング」(牧野書店)を用いて、2通りのアプローチで理解を促した。それは、従来の(1)紙と鉛筆と教科書をにらみながら問題を解く、といったスタイルの勉強に加えて、(2)統計ソフトウェア(R, EXCEL)の実行より理解する、方法である。 (2)は計算の自動化による弊害があるものの、情報集約と意思決定の簡便化とシミュレーションによる確率空間の模擬的な視覚化、実際に使用されている、という3つの大きな利点がある。

Teaching materials

  • 拙著「実例とEXCELによる統計学トレーニング」(牧野書店) G.S.Maddala「計量経済分析の方法」(シーエーピー出版)

Teaching method presentations

  • 特になし

Special notes on other educational activities

  • トライアスロンサークル「インフィニティ」の顧問(2010年秋より現在)