دانلود کتاب Research Papers in Statistical Inference for Time Series and Related Models: Essays in Honor of Masanobu Taniguchi

فهرست مطالب :

Foreword\nPreface\nPhotos of Masanobu Taniguchi\nBiography of Masanobu Taniguchi\nPublications of Masanobu Taniguchi\nStudents of Masanobu Taniguchi\nContents\nContributors\n1 Spatial Median-Based Smoothed and Self-Weighted GEL Method for Vector Autoregressive Models\n 1.1 Introduction\n 1.2 Settings\n 1.2.1 Model and Spatial Median\n 1.2.2 Self-weighted and Smoothed GEL Function\n 1.3 Main Results\n 1.4 Finite Sample Performance\n 1.5 Proofs\n 1.5.1 Some Approximations\n 1.5.2 Proofs of Theorems 1.1 and 1.2\n References\n2 Excess Mean of Power Estimator of Extreme Value Index\n 2.1 Introduction\n 2.2 Asymptotic Properties\n 2.2.1 Asymptotic Property of the Empirical Tail Process Qn(t) for Dependent Sequences\n 2.2.2 Connections Between EMP and MOP Estimators\n 2.2.3 Asymptotic Normalities of the MOP and the EMP Estimators\n 2.2.4 Consistent Estimators of σ2MOP,p,γ,r and σ2EMP,p,γ,r\n 2.3 Asymptotic Comparison of EMP with Existing Contenders for I.I.D. Observations\n 2.4 Simulations\n 2.4.1 Sensitivity Analysis of mn for mn,pEMP(kn)\n 2.4.2 Simulation Results under Various Thresholds\n 2.4.3 Simulation Results at Optimal Threshold\n 2.5 Conclusion\n 2.6 Technical Details\n 2.6.1 Details on Existence of ep(t)\n 2.6.2 Proof of Theorem 2.1\n 2.6.3 Proof of Theorem 2.2\n 2.6.4 Proof of Theorem 2.3\n 2.6.5 Proof of Theorem 2.4\n 2.6.6 Proof of Theorem 2.5\n References\n3 Exclusive Topic Model\n 3.1 Introduction\n 3.2 Method\n 3.2.1 ETM\n 3.2.2 Only Weighted LASSO Penalty: ν= 0\n 3.2.3 Only Pairwise Kullback–Leibler Divergence Penalty: µ= 0\n 3.2.4 Combination of Two Penalties\n 3.2.5 Dynamic Penalty Weight Implementation\n 3.3 Simulation\n 3.3.1 Case 1: Corpus-Specific Common Words\n 3.3.2 Case 2: Important Words Appear Rarely in Corpus\n 3.3.3 Case 3: ``Close\'\' Topics\n 3.4 Real Data Application\n 3.5 Conclusion\n References\n4 A Simple Isotropic Correlation Family in mathbbR3 with Long-Range Dependence and Flexible Smoothness\n 4.1 Introduction\n 4.1.1 A Spectral Representation\n 4.2 A New Correlation Family\n 4.3 Properties\n 4.4 Comparison With a Matérn Sub-family\n 4.5 Other Correlation Families With Long-Range Dependence\n 4.5.1 The Generalized Cauchy Family\n 4.5.2 The Confluent Hypergeometric Family\n 4.6 Discussion\n References\n5 Portmanteau Tests for Semiparametric Nonlinear Conditionally Heteroscedastic Time Series Models\n 5.1 Introduction\n 5.2 Model and Preliminaries\n 5.2.1 Asymptotic Distribution of the QML Estimator\n 5.2.2 Asymptotic Distribution of the Residuals Empirical Autocorrelations\n 5.3 Different Portmanteau Goodness-of-Fit Tests\n 5.3.1 Portmanteau Test Statistics\n 5.3.2 Bahadur Asymptotic Relative Efficiency\n 5.3.3 Nonlinear Transformations of the Residuals\n 5.4 Illustrations\n 5.4.1 Computation of Bahadur\'s Slopes in a Particular Example\n 5.4.2 Numerical Evaluation of Bahadur\'s Slopes\n 5.4.3 Monte Carlo Experiments\n 5.5 Conclusion\n 5.6 Technical Details\n References\n6 Parameter Estimation of Standard AR(1) and MA(1) Models Driven by a Non-I.I.D. Noise\n 6.1 Introduction\n 6.2 Main Results\n 6.3 Monte Carlo Experiments\n References\n7 Tests for a Structural Break for Nonnegative Integer-Valued Time Series\n 7.1 Introduction\n 7.2 Settings\n 7.3 Detection of a Structural Break\n 7.4 Numerical Study\n 7.5 Proof\n 7.5.1 Proof of Theorem 7.1\n 7.5.2 Proof of Theorem 7.2\n References\n8 M-Estimation in GARCH Models in the Absence of Higher-Order Moments\n 8.1 Introduction\n 8.2 M-Estimation of GARCH Parameters\n 8.2.1 A Class of M-Estimators\n 8.2.2 Asymptotic Distribution of M-Estimators\n 8.3 Bootstrapping M-Estimators\n 8.4 Computational Issues\n 8.4.1 Computation of the M-Estimators\n 8.4.2 Computation of the Bootstrap M-Estimators\n 8.5 Monte Carlo Comparison of Performance\n 8.5.1 GARCH(1,1) Models\n 8.5.2 GARCH(2,1) Models\n 8.5.3 A Misspecified GARCH Case\n 8.5.4 GARCH(1,2) Models\n 8.6 Performance of the Bootstrap Confidence Intervals\n 8.7 Real Data Analysis\n 8.7.1 The FTSE 100 Data\n 8.7.2 The Electric Fuel Corporation (EFCX) Data\n 8.8 Conclusion\n References\n9 Rank Tests for Randomness Against Time-Varying MA Alternative\n 9.1 Introduction\n 9.2 Preceding Studies\n 9.2.1 Linear Serial Rank Tests for White Noise Against Stationary ARMA Alternatives\n 9.2.2 Locally Asymptotic Normality of Time-Varying AR Models\n 9.3 Local Asymptotic Normality for Time-Varying MA Processes\n 9.3.1 Contiguous Hypotheses for Time-Varying MA Models\n 9.3.2 Fisher Information Matrix for Time-Varying MA Model\n 9.3.3 Central Sequence for Time-Varying MA Model\n 9.4 Main Results\n 9.4.1 Linear Serial Rank Statistics for Time-Varying MA Model\n 9.4.2 Asymptotic Normality of Test Statistics for Time-Varying MA Model\n 9.5 Concluding Remarks\n References\n10 Asymptotic Expansions for Several GEL-Based Test Statistics and Hybrid Bartlett-Type Correction with Bootstrap\n 10.1 Introduction\n 10.2 Preliminaries\n 10.2.1 Statistical Setup and GEL\n 10.2.2 Notation\n 10.2.3 Assumptions\n 10.3 Higher-Order Result of GEL Testing Inference\n 10.4 Bartlett-Type Correction with Bootstrap\n 10.5 Concluding Remarks\n 10.6 Technical Details\n 10.6.1 Some Auxiliary Lemmas\n 10.6.2 Proof of Lemma 10.1\n 10.6.3 Asymptotic Expansion for the Distribution of T(N)ρ,τ\n References\n11 An Analog of the Bickel–Rosenblatt Test for Error Density in the Linear Regression Model\n 11.1 Introduction and Summary\n 11.2 Asymptotic Null Distribution of T\"0362Tn\n 11.3 Asymptotic Distribution of T\"0362Tn Under H1\n 11.4 Random Covariates\n 11.4.1 Asymptotic Null Distribution of widetildeTn\n 11.4.2 Asymptotic Distribution of widetildeTn Under H1\n 11.5 Finite Sample Simulations and A Real Data Example\n 11.5.1 Finite Sample Simulations\n 11.5.2 Real Data Examples\n References\n12 A Minimum Contrast Estimation for Spectral Densities of Multivariate Time Series\n 12.1 Introduction\n 12.2 Contrast Function for Multivariate Time Series\n 12.3 Statistical Inference\n 12.4 Asymptotic Efficiency under Gaussianity\n 12.5 Numerical Study\n References\n13 Generalized Linear Spectral Models for Locally Stationary Processes\n 13.1 Introduction\n 13.2 A Generalized Linear Model for the Time-Varying Spectrum\n 13.2.1 Reparameterization\n 13.2.2 Modeling Structural Change: Local Generalized Cepstral Coefficients\n 13.2.3 Structural Changes Preserving Local Stationarity\n 13.3 Statistical Inference\n 13.3.1 Model Selection\n 13.4 Illustration\n 13.5 Concluding Remarks\n 13.6 Proofs\n 13.7 Proof of Theorem 13.2\n References\n14 Tango: Music, Dance and Statistical Thinking\n 14.1 Introduction\n 14.2 Dancers\' Interaction\n 14.3 Activity of the Dancers and Waiting Time\n References\n15 Z-Process Method for Change Point Problems in Time Series\n 15.1 Introduction\n 15.2 Z-process Method for Change Point Problems\n 15.3 Change Point Problems for Time Series\n 15.4 Nonlinear Time Series Models\n References\n16 Copula Bounds for Circular Data\n 16.1 Introduction\n 16.2 Equivalence Class of Circular Copula Functions\n 16.3 Circular Fréchet–Hoeffding Copula Bounds\n 16.4 Monte Carlo Simulations\n 16.5 Summary and Conclusions\n 16.6 Proof of Theorem 16.1\n References\n17 Topological Data Analysis for Directed Dependence Networks of Multivariate Time Series Data\n 17.1 Introduction\n 17.2 VAR Models and PDC\n 17.3 Network Decomposition\n 17.4 Overview of Persistent Homology and Vietoris–Rips Filtration\n 17.5 Oriented TDA of Seizure EEG\n 17.6 Conclusion\n References\n18 Orthogonal Impulse Response Analysis in Presence of Time-Varying Covariance\n 18.1 Introduction\n 18.2 Time-Varying Orthogonal Impulse Response Functions\n 18.2.1 tv-OIRFs\n 18.2.2 Approximated OIRFs\n 18.2.3 Averaged OIRFs\n 18.2.4 Variance Variability Indices\n 18.3 OIRFs Estimators When the Variance is Varying\n 18.3.1 The tv-OIRFs Nonparametric Estimator\n 18.3.2 Approximated OIRFs Estimators\n 18.3.3 New OIRFs Estimators With tv-Variance\n 18.4 Conclusion\n 18.5 Technical Details\n 18.5.1 Kernel Estimators of the Covariance Function Integrals\n 18.5.2 Assumptions\n 18.5.3 Proofs\n References\n19 Robustness Aspects of Optimal Transport\n 19.1 Introduction\n 19.2 Optimal Transport\n 19.2.1 Basic Formulation\n 19.2.2 Wasserstein Distance\n 19.3 Robust Approaches\n 19.3.1 Robustifying Estimators via Wasserstein Distance\n 19.3.2 Saturation in Linear Models\n 19.4 Robust Optimal Transport\n 19.5 Related Techniques and Outlook\n References\n20 Estimating Finite-Time Ruin Probability of Surplus with Long Memory via Malliavin Calculus\n 20.1 Introduction\n 20.2 Preliminaries\n 20.2.1 Notation\n 20.2.2 Malliavin Calculus\n 20.3 Statistical Problems\n 20.3.1 Estimation of σ\n 20.3.2 Simulation-Based Inference for Ψσ(u, T)\n 20.4 Differentiability of Ψσ\n References\n21 Complex-Valued Time Series Models and Their Relations to Directional Statistics\n 21.1 Introduction\n 21.2 Complex-Valued Time Series\n 21.2.1 Wrapped Cauchy Process\n 21.2.2 von Mises Process\n 21.2.3 Sine-Skewed Process\n 21.2.4 Random Walks\n 21.3 Parameter Estimation\n 21.3.1 Method of Moments Estimation\n 21.3.2 Whittle Estimation\n 21.4 Monte Carlo Simulations\n 21.5 Data Analysis\n 21.5.1 Old Faithful Geyser Data\n 21.5.2 Real Wage and Unemployment Rate in Canada\n 21.6 Summary and Conclusions\n References\n22 Semiparametric Estimation of Optimal Dividend Barrier for Spectrally Negative Lévy Process\n 22.1 Introduction\n 22.2 Optimal Dividend Barrier\n 22.3 Estimation of Optimal Dividend Barrier\n 22.4 Asymptotic Results\n 22.5 Numerical Results\n 22.5.1 Quasi-process\n 22.5.2 Maximum Contrast Estimator\n 22.5.3 Simulation Result\n 22.6 Proofs\n 22.6.1 Proof of Lemma 22.1\n 22.6.2 Proof of Lemma 22.2\n 22.6.3 Proof of Lemma 22.3\n 22.6.4 Proof of Lemma 22.4\n 22.6.5 Proof of Theorem 22.2\n References\n23 Local Signal Detection for Categorical Time Series\n 23.1 Introduction\n 23.2 Spectral Envelope\n 23.2.1 Estimation\n 23.2.2 An Example\n 23.3 Local Analysis\n 23.3.1 Local Whittle Likelihood\n 23.3.2 Minimum Description Length\n 23.3.3 Optimization via Genetic Algorithm\n 23.3.4 Another Example\n References\n24 Topological Inference on Electroencephalography\n 24.1 Introduction\n 24.2 Preliminary\n 24.3 Methods\n 24.3.1 Persistent Homology on a Signal\n 24.3.2 Permutation-Based Topological Inference\n 24.4 Simulations\n 24.5 Applications\n 24.6 Discussion\n References\n25 UMVU Estimation for Time Series\n 25.1 Introduction\n 25.2 Methodology\n 25.3 Tests by Monotone Likelihood Ratio\n 25.4 Simulation\n 25.5 Conclusion\n References\n26 A New Generalized Estimator for AR(1) Model Which Improves MLE Uniformly\n 26.1 Introduction\n 26.2 A Generalized Estimator of the AR Coefficient\n 26.3 A New Estimator Which Improves MLE Uniformly\n 26.4 Numerical Studies\n References