ISBN: 3540675450
TITLE: XploRe
AUTHOR: Hrdle, Wolfgang; Hlavka, Zdenek; Klinke, Sigbert
TOC:

I Regression Models 17
1 Quantile Regression 19
Pavel Czek
1.1 Introduction 19
1.2 Quantile Regression 22
1.2.1 Definitions 23
1.2.2 Computation 26
1.3 Essential Properties 28
1.3.1 Equivariance 28
1.3.2 Invariance to Transformations 28
1.3.3 Robustness 29
1.4 Inference 32
1.4.1 Main Asymptotic Results 33
1.4.2 Wald Test 34
1.4.3 Rank Tests 36
1.5 Description of Quantlets 41
1.5.1 Quantlet rqfit 41
1.5.2 Quantlet rrstest 45
Bibliography 47
2 Least Trimmed Squares 49
Pavel Czek and Jan mos Vek
2.1 Robust Regression 49
2.1.1 Introduction 49
2.1.2 High Breakdown point Estimators 52
2.2 Least Trimmed Squares 54
2.2.1 Definition 54
2.2.2 Computation 56
2.3 Supplementary Remarks 58
2.3.1 Choice of the Trimming Constant 58
2.3.2 LTS as a Diagnostic Tool 59
2.3.3 High Subsample Sensitivity 60
Bibliography 62
3 Errors-in-Variables Models 65
Hua Liang
3.1 Linear EIV Models 65
3.1.1 A Single Explanatory Variable 67
3.1.2 Vector of Explanatory Variables 78
3.2 Nonlinear EIV Models 83
3.2.1 Regression Calibration 85
3.2.2 Simulation Extrapolation 87
3.3 Partially Linear EIV Models 89
3.3.1 The Variance of Error Known 89
3.3.2 The Variance of Error Unknown 90
3.3.3 XploRe Calculation and Practical Data 91
Bibliography 95
4 Simultaneuos-Equations Models 97
Axel Werwatz and Christian Mller
4.1 Introduction 97
4.2 Estimation 98
4.2.1 Identification 98
4.2.2 Some Notation 99
4.2.3 Two-Stage Least Squares 100
4.2.4 Three-Stage Least Squares 101
4.2.5 Computation 103
4.3 Application: Money-Demand 108
Bibliography 114
5 Hazard Regression 115
Birgit Grund and Lijian Yang
5.1 Data Structure 116
5.2 Kaplan-Meier Estimates 121
5.3 The Cox Proportional Hazards Model 126
5.3.1 Estimating the Regression Coecients 127
5.3.2 Estimating the Hazard and Survival Functions 131
5.3.3 Hypothesis Testing 137
5.3.4 Example: Length of Stay in Nursing Homes 141
Bibliography 143
6 Generalized Partial Linear Models 145
Marlene Mller
6.1 Estimating GPLMs 145
6.1.1 Models 146
6.1.2 Semiparametric Likelihood 146
6.2 Data Preparation 150
6.2.1 General 150
6.2.2 Example 151
6.3 Computing GPLM Estimates 152
6.3.1 Estimation 153
6.3.2 Estimation in Expert Mode 156
6.4 Options 158
6.4.1 Setting Options 159
6.4.2 Grid and Starting Values 159
6.4.3 Weights and Offsets 160
6.4.4 Control Parameters 161
6.4.5 Model Parameters 162
6.4.6 Specification Test 162
6.4.7 Output Modification 162
6.5 Statistical Evaluation and Presentation 163
6.5.1 Statistical Characteristics 163
6.5.2 Output Display 164
6.5.3 Model selection 165
7 Generalized Additive Models 171
Stefan Sperlich and Jir Zelinka
7.1 Brief Theory 172
7.1.1 Models 172
7.1.2 Marginal Integration 173
7.1.3 Backfitting 174
7.1.4 Orthogonal Series 175
7.2 Data Preparation 176
7.3 Noninteractive Quantlets for Estimation 176
7.3.1 Estimating an AM 177
7.3.2 Estimating an APLM 179
7.3.3 Estimating an AM and APLM 181
7.3.4 Estimating a GAM 184
7.3.5 Estimating a GAPLM 186
7.3.6 Estimating Bivariate Marginal In uence 189
7.3.7 Estimating an AM with Interaction Terms 191
7.3.8 Estimating an AM Using Marginal Integration 195
7.4 Interactive Quantlet GAMFIT 197
7.5 How to Append Optional Parameters 203
7.6 Noninteractive Quantlets for Testing 205
7.6.1 Component Analysis in APL Models 206
7.6.2 Testing for Interaction 208
7.6.3 Testing for Interaction 210
7.7 Odds and Ends 213
7.7.1 Special Properties of GAM Quantlib Quantlets 213
7.7.2 Estimation on Principal Component by PCAD 213
7.8 Application for Real Data 215
Bibliography 220
II Data Exploration 221
8 Growth Regression and Counterfactual Income Dynamics 223
Alain Desdoigts
8.1 A Linear Convergence Equation 224
8.2 Counterfactual Income Dynamics 226
8.2.1 Sources of the Growth Differential With Respect to a
Hypothetical Average Economy 226
8.2.2 Univariate Kernel Density Estimation and Bandwidth
Selection 227
8.2.3 Multivariate Kernel Density Estimation 234
9 Cluster Analysis 239
Hans-Joachim Mucha and Hizir Sofyan
9.1 Introduction 239
9.1.1 Distance Measures 240
9.1.2 Similarity of Objects 242
9.2 Hierarchical Clustering 243
9.2.1 Agglomerative Hierarchical Methods 244
9.2.2 Divisive Hierarchical Methods 258
9.3 Nonhierarchical Clustering 262
9.3.1 K-means Method 263
9.3.2 Adaptive K-means Method 265
9.3.3 Hard C-means Method 267
9.3.4 Fuzzy C-means Method 269
Bibliography 278
10 Classification and Regression Trees 281
Jussi Klemel, Sigbert Klinke, and Hizir Sofyan
10.1 Growing the Tree 281
10.2 Pruning the Tree 283
10.3 Selecting the Final Tree 284
10.4 Plotting the Result of CART 285
10.5 Examples 287
10.5.1 Simulated Example 287
10.5.2 Boston Housing Data 290
10.5.3 Density Estimation 296
11 DPLS: Partial Least Squares Program 305
Frank Geppert and Hans Gerhard Strohe
11.1 Introduction 305
11.2 Theoretical Background 307
11.2.1 The Dynamic Path Model DPLS 307
11.2.2 PLS Estimation with Dynamic Inner Approximation 307
11.2.3 Prediction and Goodness of Fit 309
11.3 Estimating a DPLS-Model 311
11.3.1 The Computer Program DPLS 311
11.3.2 Creating design-matrices 311
11.3.3 Estimating with DPLS 314
11.3.4 Measuring the Forecasting Validity 317
11.4 Example: A Model for German Share Prices 318
11.4.1 The General Path Model 318
11.4.2 Manifest Variables and Sources of Data 319
11.4.3 Empirical Results 321
Bibliography 322
12 Uncovered Interest Parity 323
Jrg Breitung and Ralf Brggemann
12.1 The Uncovered Interest Parity 323
12.2 The Data 325
12.3 A Fixed Effects Model 327
12.4 A Dynamic Panel Data Model 330
12.5 Unit Root Tests for Panel Data 332
12.6 Conclusions 335
12.7 Macro Data 336
Bibliography 336
13 Correspondence Analysis 339
Michal Benko and Michel Lejeune
13.1 Introduction 339
13.1.1 Singular Value Decomposition 339
13.1.2 Coordinates of Factors 340
13.2 XploRe Implementation 341
13.3 Example: Eye-Hair 341
13.3.1 Description of Data 341
13.3.2 Calling the Quantlet 342
13.3.3 Documentation of Results 343
13.3.4 Eigenvalues 343
13.3.5 Contributions 343
13.3.6 Biplots 347
13.3.7 Brief Remark 350
13.4 Example: Media 350
13.4.1 Description of the Data Set 350
13.4.2 Calling the Quantlet 352
13.4.3 Brief Interpretation 352
Bibliography 358
III Dynamic Statistical Systems 359
14 Long-Memory Analysis 361
Gilles Teyssiere
14.1 Introduction 361
14.2 Model Indepependent Tests for I(0) against I(d) 363
14.2.1 Robust Rescaled Range Statistic 364
14.2.2 The KPSS Statistic 366
14.2.3 The Rescaled Variance V/S Statistic 367
14.2.4 Nonparametric Test for I(0) 369
14.3 Semiparametric Estimators in the Spectral Domain 371
14.3.1 Log-periodogram Regression 371
14.3.2 Semiparametric Gaussian Estimator 373
Bibliography 374
15 ExploRing Persistence in Financial Time Series 377
David Lee
15.1 Introduction 377
15.2 Hurst and Fractional Integration 379
15.2.1 Hurst Constant 379
15.2.2 Fractional Integration 379
15.3 Tests for I(0) against fractional alternatives 380
15.4 Semiparametric estimation of difference parameter d 380
15.5 ExploRing the Data 381
15.5.1 Typical Spectral Shape 381
15.5.2 Typical Distribution: Mean, Variance, Skewness and Kurtosis 383
15.6 The Data 384
15.7 The Quantlets 386
15.8 The Results 388
15.8.1 Equities 388
15.8.2 Exchange 389
15.9 Practical Considerations 390
15.9.1 Risk and Volatility 390
15.9.2 Estimating and Forecasting of Asset Prices 391
15.9.3 Portfolio Allocation Strategy 391
15.9.4 Diversification and Fractional Cointegration 392
15.9.5 MMAR and FIGARCH 392
15.10 Conclusion 393
Bibliography 393
16 Flexible Time Series Analysis 397
Wolfgang Hrdle and Rolf Tschernig
16.1 Nonlinear Autoregressive Models of Order One 398
16.1.1 Estimation of the Conditional Mean 398
16.1.2 Bandwidth Selection 407
16.1.3 Diagnostics 410
16.1.4 Confidence Intervals 413
16.1.5 Derivative Estimation 417
16.2 Nonlinear Autoregressive Models of Higher Order 420
16.2.1 Estimation of the Conditional Mean 421
16.2.2 Bandwidth and Lag Selection 427
16.2.3 Plotting and Diagnostics 437
16.2.4 Estimation of the Conditional Volatility 440
17 Multiple Time Series Analysis 459
Alexander Benkwitz
17.1 Getting Started 459
17.1.1 Data Preparation 460
17.1.2 Starting multi 460
17.2 Preliminary Analysis 461
17.2.1 Plotting the Data 462
17.2.2 Data Transformation 463
17.3 Specifying a VAR Model 465
17.3.1 Process Order 466
17.3.2 Model Estimation 468
17.3.3 Model Validation 470
17.4 Structural Analysis 477
17.4.1 Impulse Response Analysis 477
17.4.2 Confidence Intervals for Impulse Responses 480
18 Robust Kalman Filtering 483
Peter Ruckdeschel
18.1 State-Space Models and Outliers 483
18.1.1 Outliers and Robustness Problems 484
18.1.2 Examples of AO's and IO's 488
18.1.3 Problem Setup 490
18.2 Classical Method: Kalman Filter 491
18.2.1 Features of the Classical Kalman Filter 491
18.2.2 Optimality of the Kalman Filter 492
18.3 The rLS filter 493
18.3.1 Derivation 493
18.3.2 Calibration 494
18.3.3 Examples 495
18.3.4 Possible Extensions 502
18.4 The rIC filter 502
18.4.1 Filtering = Regression 504
18.4.2 Robust Regression Estimates 504
18.4.3 Variants: Separate Clipping 506
18.4.4 Criterion for the Choice of b 506
18.4.5 Examples 506
18.4.6 Possible Extensions 510
18.5 Generating In uence Curves 512
18.5.1 Definition of IC 512
18.5.2 General Algorithm 513
18.5.3 Explicite Calculations 514
18.5.4 Integrating along the Directions 514
18.5.5 Auxiliary routines 515
Bibliography 516
Index 517
END
