Classification and Data Analysis : Theory and Applications

This volume gathers peer-reviewed contributions on data analysis, classification and related areas presented at the 28th Conference of the Section on Classification and Data Analysis of the Polish Statistical Association, SKAD 2019, held in Szczecin, Poland, on September 18–20, 2019. Providing a balance between theoretical and methodological contributions and empirical papers, it covers a broad variety of topics, ranging from multivariate data analysis, classification and regression, symbolic (and other) data analysis, visualization, data mining, and computer methods to composite measures, and numerous applications of data analysis methods in economics, finance and other social sciences. The book is intended for a wide audience, including researchers at universities and research institutions, graduate and doctoral students, practitioners, data scientists and employees in public statistical institutions. Preface 6 Contents 9 About the Editors 12 Methods 13 Comparison of Proposals of Transformation of Nominants into Stimulants on the Example of Financial Ratios of Companies Listed on the Warsaw Stock Exchange 14 1 Introduction 15 2 Methodology 16 2.1 Justification of the Choice of Indicators-Nominants with the Recommended Range of Values 16 2.2 Selected Formulas of Transformation of Nominants with the Recommended Ranges of Values into Stimulants with the Range of [0; 1] 17 2.3 The Authors’ Proposal of the Modification of Transformation of Nominants with the Recommended Range of Values into Stimulants 21 3 Application of Examined Transformations for Selected Indicators-Nominants—Empirical Example 23 4 Conclusions 25 References 27 Silhouette Index as Clustering Evaluation Tool 29 1 Introduction 29 2 Finding Number of Clusters with Silhouette Index 30 3 Clustering Methods Based on Silhouette Index 34 4 The Performance of Clustering Methods Based on Silhouette Index 36 5 Conclusions and Remarks 39 Appendix 1: Source Code of Procedure of Finding Number of Cluster with Silhouette Index in R Language 41 Appendix 2: Source Code of Clustering Tree (RPCT) Algorithm Implementation in R Language 42 References 43 The Role of Discretization of Continuous Variables in Socioeconomic Classification Models on the Example of Logistic Regression Models and Artificial Neural Networks 44 1 Introduction 45 2 Discretization of Continuous Variables 47 3 Empirical Examples 50 4 Summary and Conclusions 58 References 59 Intuitionistic Fuzzy Synthetic Measure for Ordinal Data 61 1 Introduction 61 2 Fuzzy Sets Theory 62 3 Intuitionistic Fuzzy Sets Theory 63 4 Transformation of Ordinal Data into Intuitionistic Fuzzy Sets 64 5 Intuitionistic Fuzzy Synthetic Measure 65 6 Empirical Example 66 7 A Comparative Analysis 69 8 Conclusions 79 References 80 Improving Classification Accuracy of Ensemble Learning for Symbolic Data Trough Neural Networks’ Feature Extraction 81 1 Introduction 81 2 Symbolic Data and Variable Selection from Neural Networks 82 3 Results of Credit Scoring 87 4 Final Remarks 90 References 91 Inequality Restricted Least Squares (IRLS) Model of Real Estate Prices 94 1 Introduction 94 2 Literature Review 96 3 Research Methodology 97 4 Data and Research Results 100 5 Conclusions 106 References 107 Application of Hill Estimator to Assess Extreme Risks in the Metals Market 109 1 Introduction 109 2 Extreme Risk 110 3 Hill Estimator and Its Modification 111 4 Empirical Study 113 5 Conclusions 118 References 118 Segmentation of Enterprises on the Basis of Their Duration Using Survival Trees—Results of an Analysis for Legal Persons and Organizational Entities Without Legal Personality in the Łódzkie Voivodship 120 1 Introduction 121 2 Data Characteristics and Methods 123 3 Results 126 4 Conclusions 131 References 132 Corporate Bankruptcy Prediction with the Use of the Logit Leaf Model 134 1 Introduction 134 2 Data and Research Procedure 136 3 Empirical Results for the Balanced Sets 138 4 Empirical Results for the Non-balanced Sets 144 5 Conclusions 147 References 150 The Impact of Longevity on a Valuation of Long-Term Investments Returns: The Case of Selected European Countries 152 1 Introduction 152 2 Longevity and Financial Markets 153 2.1 Historical Trends and Projections Review 153 2.2 The Implication of Longevity and Aging on Investments Returns 154 3 Methodology 155 4 Results 158 5 Conclusions 164 References 164 Applications in Economics 166 Sustainable Development and Green Economy in the European Union Countries—Statistical Analysis 167 1 Introduction 168 2 Literature Review 169 3 Procedure of the Conducted Research 171 3.1 Sources of Information on Diagnostic Indicators 171 3.2 Selection of Diagnostic Features 172 3.3 Research Method 176 4 The Study Results 177 5 Discussion and Conclusions 185 References 187 The Review of Indicators of Data Quality in Intra-Community Trade in Goods. The Choice of an Indicator and Its Effect on the Ranking of Countries 190 1 Introduction 190 2 Literature Review 191 3 Proposed Methods 195 4 Results 197 5 Conclusions 202 References 203 Development of ICT in Poland in Comparison with the European Union Countries—Multivariate Statistical Analysis 205 1 Introduction 206 2 Data Sources and Methods 207 3 Results 208 3.1 The Assessment of the ICT Access Changes in the Households in Years 2008–2017 208 3.2 The Analysis of the Lack of Internet Access Reasons in Years 2010–2017 213 3.3 The Analysis of the Internet Activities in Years 2010–2017 216 4 Conclusions 219 References 221 Sensitivity Analysis in Causal Mediation Effects for TAM Model 223 1 The Concept of Causality 223 2 Technology Acceptance Model 224 3 Identification of Causal Effects 225 4 Mediation Effects in TAM 227 5 D-Separation of Causal Relationships in TAM 230 6 Sensitivity Analysis of TAM 231 7 Conclusions 234 References 235 Prentice–Williams–Peterson Models in the Assessment of the Influence of the Characteristics of the Unemployed on the Intensity of Subsequent Registrations in the Labour Office 238 1 Introduction 238 2 Multiple Registrations with the Employment Office 240 3 Research Methodology 241 4 Data Used in the Study 244 5 Analysis of Multiple Registrations in the Labour Office 245 6 Conclusions 248 References 250 Right-Skewed Distribution of Features and the Identification Problem of the Financial Autonomy of Local Administrative Units 252 1 Introduction 253 2 Research Methods 254 2.1 Source Material 254 2.2 The Methods 255 3 Results 259 3.1 Characteristics of the Simple Features 259 3.2 Identification of Financial Autonomy Levels of Polish Communes 260 4 Conclusions 262 References 264 Multi-criteria Rankings with Interdependent Criteria: Case of EU Countries on Their Way to Healthy Lives and Well-Being 266 1 Introduction 266 2 Some Methodological Issues 267 2.1 Normalization 267 2.2 Weighting Using Informative Capacity of the Criteria and PCA Approach 268 2.3 Multi-criteria Principal Components (MCPC) Method 271 3 Results of EU Countries Comparisons with Respect to “Healthy Lives and Well-Being” 272 4 Conclusions 284 References 288 The Comparison of Income Distributions for Women and Men in the European Union Countries 290 1 Introduction 290 2 Research Methodology 292 3 Empirical Data and Results 293 4 Conclusions 301 References 303 Common Stochastic Mortality Trends for Multiple European Populations 305 1 Introduction 305 2 Mortality Modeling 306 3 Identification of Common Stochastic Mortality Trend 307 3.1 Spatial Clustering Method 307 3.2 Semi-parametric Method 308 4 Common Mortality Trends in Europe 311 4.1 European Mortality Trends 311 4.2 Two-Country Mortality Case 312 4.3 European Common Low-Mortality Trend 314 5 Conclusions 316 References 317 Impact of the Selected Factors on the Men and Women Wages in Poland in 2014. The Conjoint Analysis Application 318 1 Introduction 318 2 Background for the Wages Analysis 319 3 Data and Applied Methods 321 4 Results 326 5 Conclusions 331 References 332 Front Matter ....Pages i-xiii Front Matter ....Pages 1-1 Comparison of Proposals of Transformation of Nominants into Stimulants on the Example of Financial Ratios of Companies Listed on the Warsaw Stock Exchange (Barbara Batóg, Katarzyna Wawrzyniak)....Pages 3-17 Silhouette Index as Clustering Evaluation Tool (Andrzej Dudek)....Pages 19-33 The Role of Discretization of Continuous Variables in Socioeconomic Classification Models on the Example of Logistic Regression Models and Artificial Neural Networks (Wioletta Grzenda)....Pages 35-51 Intuitionistic Fuzzy Synthetic Measure for Ordinal Data (Bartłomiej Jefmański)....Pages 53-72 Improving Classification Accuracy of Ensemble Learning for Symbolic Data Trough Neural Networks’ Feature Extraction (Marcin Pełka)....Pages 73-84 Front Matter ....Pages 85-85 Inequality Restricted Least Squares (IRLS) Model of Real Estate Prices (Mariusz Doszyń)....Pages 87-101 Application of Hill Estimator to Assess Extreme Risks in the Metals Market (Dominik Krężołek)....Pages 103-113 Segmentation of Enterprises on the Basis of Their Duration Using Survival Trees—Results of an Analysis for Legal Persons and Organizational Entities Without Legal Personality in the Łódzkie Voivodship (Artur Mikulec, Małgorzata Misztal)....Pages 115-128 Corporate Bankruptcy Prediction with the Use of the Logit Leaf Model (Barbara Pawełek, Józef Pociecha)....Pages 129-146 The Impact of Longevity on a Valuation of Long-Term Investments Returns: The Case of Selected European Countries (Grażyna Trzpiot)....Pages 147-160 Front Matter ....Pages 161-161 Sustainable Development and Green Economy in the European Union Countries—Statistical Analysis (Katarzyna Cheba, Iwona Bąk)....Pages 163-185 The Review of Indicators of Data Quality in Intra-Community Trade in Goods. The Choice of an Indicator and Its Effect on the Ranking of Countries (Iwona Markowicz, Paweł Baran)....Pages 187-201 Development of ICT in Poland in Comparison with the European Union Countries—Multivariate Statistical Analysis (Małgorzata Misztal, Aleksandra Kupis-Fijałkowska)....Pages 203-220 Sensitivity Analysis in Causal Mediation Effects for TAM Model (Adam Sagan, Mariusz Grabowski)....Pages 221-234 Front Matter ....Pages 235-235 Prentice–Williams–Peterson Models in the Assessment of the Influence of the Characteristics of the Unemployed on the Intensity of Subsequent Registrations in the Labour Office (Beata Bieszk-Stolorz)....Pages 237-250 Right-Skewed Distribution of Features and the Identification Problem of the Financial Autonomy of Local Administrative Units (Romana Głowicka-Wołoszyn, Feliks Wysocki)....Pages 251-264 Multi-criteria Rankings with Interdependent Criteria: Case of EU Countries on Their Way to Healthy Lives and Well-Being (Iwona Konarzewska)....Pages 265-288 The Comparison of Income Distributions for Women and Men in the European Union Countries (Joanna Landmesser)....Pages 289-303 Common Stochastic Mortality Trends for Multiple European Populations (Justyna Majewska, Grażyna Trzpiot)....Pages 305-317 Impact of the Selected Factors on the Men and Women Wages in Poland in 2014. The Conjoint Analysis Application (Aleksandra Matuszewska-Janica)....Pages 319-335