Machine-learning Techniques in Economics (eBook)

Contents 6
Chapter 1: Why This Book? 8
References 13
Chapter 2: Data, Variables, and Their Sources 14
2.1 Variables and Their Sources 19
2.2 Problems with Institutional Measures 22
2.3 Imputing Missing Data 25
References 25
Chapter 3: Methodology 26
3.1 Estimation Techniques 27
3.1.1 Artificial Neural Networks 28
3.1.2 Regression Tree Predictors 29
3.1.3 Boosting Algorithms 30
3.1.4 Bootstrap Aggregating (Bagging) Predictor 31
3.1.5 Random Forests 32
3.2 Predictive Accuracy 33
3.3 Variable Importance and Partial Dependence 34
References 35
Chapter 4: Predicting a Country´s Growth: A First Look 36
References 43
Chapter 5: Predicting Economic Growth: Which Variables Matter 44
5.1 Evaluating Traditional Variables 47
5.2 Policy Levers 52
References 62
Chapter 6: Predicting Recessions: What We Learn from Widening the Goalposts 64
6.1 Predictive Quality 65
6.2 Variable Importance and Partial Dependence Plots: What Do We Learn? 69
6.2.1 The First Lens: Implications for Modeling Recessions Theoretically 69
6.2.2 The Second Lens: A Policy Maker and a Data Scientist Walk into a Bar 72
References 80
Epilogue 81
Appendix: R Codes and Notes 83
Imputing and Processing the Data 83
Training the Models 85
Evaluating Predictive Quality 88
Variable Importance and Partial Dependence 93
References 97