Suchen und Finden
Service
Forecasting International Migration in Europe: A Bayesian View
Jakub Bijak
Verlag Springer-Verlag, 2010
ISBN 9789048188970 , 318 Seiten
Format PDF, OL
Kopierschutz Wasserzeichen
Acknowledgements
7
Contents
9
About the Author
15
About the Contributor
16
List of Figures
17
List of Tables
19
Part I Introduction
22
1 Introduction and Background
23
1.1 Migration Forecasting as a Research Problem
23
1.1.1 Role of International Migration and Its Forecasts
23
1.1.2 Problems with Errors in Migration Forecasting
25
1.2 Aim and Structure of the Book
26
1.2.1 Aim and Scope
26
1.2.2 Structure of the Book
27
1.3 Terminology and Symbols
30
1.3.1 Basic Terms Used in the Study
30
1.3.2 Mathematical Notation
31
1.3.3 Bibliographical Notation
32
2 Preliminaries
34
2.1 Definitions and Measurement of International Migration
34
2.1.1 Data Sources and Definitions
34
2.1.2 Quality and Comparability of Migration Data
35
2.1.3 Ways of Dealing with Deficient Statistics
36
2.2 Uncertainty, Subjectivity and Judgement in Population and Migration Forecasting
39
2.2.1 Uncertainty in Demographic and Migration Forecasting
42
2.2.2 Subjectivity and Judgement in Population and Migration Predictions
45
2.3 Bayesian Inference in Statistics: Introductory Notes
46
2.3.1 The Bayesian Paradigm
46
2.3.2 Decision-Theory, Empirical and Orthodox Approaches
48
2.3.3 Bayesian Interval Estimation and Forecasting
50
2.4 Markov Chain Monte Carlo (MCMC) Simulations
51
2.4.1 Numerical Solutions to Bayesian Problems: General Remarks
51
2.4.2 Simulation of Posterior Distributions Using Gibbs Sampling
52
Part II Explaining and Forecasting Migration
54
3 Explaining Migration: Brief Overview of Selected Theories
55
3.1 Different Theoretical Perspectives on Migration Flows
55
3.1.1 Sociological Theories
55
3.1.2 Macroeconomic Theories
58
3.1.3 Microeconomic Theories
61
3.1.4 Geographical Theories
63
3.1.5 Unifying Perspectives
64
3.2 Theory in Migration Forecasting: A Global Outlook
65
3.2.1 Migration Theories: General Remarks
65
3.2.2 Use of Theories for Migration Predictions
68
4 Forecasting Migration: Selected Models and Methods
70
4.1 Deterministic Methods of Predicting Migration
70
4.1.1 Judgemental Migration Scenarios
70
4.1.2 The Delphi Method and Surveys Among Experts
72
4.1.3 'Migration Potential' Assessment Surveys
73
4.1.4 Macro-Level Mathematical Models in Demography
75
4.1.5 Demo-Economic Modelling Attempts
77
4.2 Probabilistic Migration Forecasts: Assessing Uncertainty
78
4.2.1 Markovian and Related Models of Aggregate Population Flows
78
4.2.2 Micro-Level Methods: Event-History Analysis and Ethnosurvey
81
4.2.3 Selected Attempts to Bridge the Micro and Macro Perspectives
84
4.2.4 Econometric Forecasts of International Migration
85
4.2.5 Limitations of Econometric Models
89
4.2.6 Stochastic Forecasts of Migration Time Series
91
4.3 Bayesian Approach in Migration Studies and Demography
94
4.3.1 Bayesian Models and Forecasts of Population Flows
94
4.3.2 Bayesian Methods in Demography: A Concise Survey
95
4.4 From Migration Theories to Model-Based Forecasting
96
4.4.1 Migration Forecasting Methods and Models: State of the Art and Typology
96
4.4.2 Deterministic Character of Many Existing Predictions
98
4.4.3 Notes on Including Theory in Population and Migration Forecasts
99
4.4.4 Implications for the Current and Future Studies
101
Part III Examples of Bayesian Migration Predictions
105
5 Bayesian Model Selection and Forecast Averaging
106
5.1 Selection and Averaging Problems: Simple Stochastic Processes
106
5.1.1 Methodological Foundations of Bayesian Model Selection
106
5.1.2 Bayesian Forecast Averaging (Inference Pooling)
108
5.1.3 Empirical Application: Specification of Forecasting Models
110
5.1.4 Computations: The Carlin--Chib Algorithm
112
5.2 Simple Time Series Forecasts: Individual and Averaged
113
5.2.1 Estimation of the Models and Calculation of Their Posterior Probabilities
113
5.2.2 Predictions Based on the Formally-Selected and Averaged Stochastic Processes
121
5.2.3 Interpretation of Forecasts and the Comparison of Ex-Post Errors for 2005--2007
128
6 Bayesian VAR Modelling x2018;from General to Specificx2019
132
6.1 VAR Processes and Lindleys Tests for Restrictions
132
6.1.1 Selection of Predictors in Econometric Models: Rationale for the VAR Modelling
132
6.1.2 VAR Models and the 'from General to Specific' Approach
133
6.1.3 Inference on the Impact of Additional Variables on Migration
136
6.2 Example: Migration Forecasts from General and Reduced VARs
138
6.2.1 Applying the Reduction Approach: Model Specification, Estimation, and Testing
138
6.2.2 Results of Forecasts from the General VAR and Marginal AR Models
144
6.2.3 'From General to Specific' Modelling: Discussion of the Outcomes
149
7 Selected Approaches to Discontinuities in Trends
152
7.1 From Deterministic Analogies to Stochastic Volatility
152
7.1.1 The Simplest Options: Dummy Variables and Forecasting by Analogy
152
7.1.2 Models with Changing Conditional Variance ARCH, GARCH, Stochastic Volatility
154
7.2 Example: Forecasts from Models with Discontinuities
155
7.2.1 Application to Polish-German Flows: Models with Analogy to Iberian Migration
155
7.2.2 Models with Changing Conditional Variance: Model Selection for AR(1) Extensions
159
7.2.3 Predictions Prepared with Models Acknowledging Discontinuity in Trends
162
8 Evaluation of Presented Forecasts of European Migration
168
8.1 Robustness of Forecasts Against Certain Changes In Priors
168
8.1.1 Role of Sensitivity Analysis in the Bayesian Approach: Basic Remarks
168
8.1.2 Robustness of Forecasts Yielded by Selected Models
170
8.1.3 Discussion and Tentative Conclusions
175
8.2 Comparison of Selected Bayesian and Frequentist Forecasts
176
8.2.1 General Remarks on Ex-Ante and Ex-Post Prediction Errors
176
8.2.2 Likelihood-Based Estimation and Model Selection Framework
177
8.2.3 Selected Bayesian and Frequentist Migration Forecasts for 2000--2007
180
8.2.4 Comparison of Ex-Ante and Ex-Post Errors for Various Predictions
184
9 Bayesian Computing in Practice
189
9.1 A Short Survey of Available Bayesian Software
189
9.1.1 R programming Language
189
9.1.2 Octave
190
9.1.3 BUGS
191
9.2 Bayesian Computation in WinBUGS
191
9.2.1 Model and Data Specification
191
9.2.2 Model Compilation, Initialisation and Updating
193
9.2.3 Convergence Diagnostics and Inference
194
9.3 Example of Bayesian Computation in R Language
195
9.3.1 Forecasting Migration Using R
195
9.3.2 The Model of Immigration Flows
195
9.3.3 Sampling
197
9.3.4 Carlin--Chib Model Selection Procedure
203
9.4 Conclusions
209
Part IV Perspectives of Forecast Makers and Users
211
10 Extensions and Limitations of Migration Forecasts
212
10.1 Data, Theories and Judgement: Towards a Synthesis?
212
10.1.1 Theory in an Atheoretical Setting: Prior Distributions in Multivariate Models
212
10.1.2 Data Versus Judgement: Elicitation of Expert Knowledge
215
10.2 Controlling Plausibility of Outcomes in Demographic Models
218
10.2.1 Combining Deterministic Population Models with Stochastic Forecasts
218
10.2.2 The Bayesian Melding Approach: Outline and Discussion
220
10.3 Imperfect Knowledge Forecasting of Migration and Population
221
10.3.1 Micro-level Foundations in Macro-level Forecasting
221
10.3.2 The Imperfect Knowledge Paradigm: Quantitative Versus Qualitative Predictions
222
10.4 Implications for Forecast-Makers and Future Research Agenda
224
10.4.1 Limitations of Predictability and Plausible Horizon of Non-stationary Forecasts
224
10.4.2 Forecasting Migration and Population: Proposal for a Research Agenda
227
11 Dealing with Uncertain Forecasts: A Policy Perspective
230
11.1 Preliminaries of the Decision Analysis: A Bayesian Perspective
230
11.1.1 Background: Selected Insights into Decisions and Attitudes Towards Uncertainty
230
11.1.2 Estimation and Prediction in the Bayesian Decision Framework
233
11.1.3 Bayesian Decision Analysis: Some Stylised Examples
236
11.1.4 Possible Extensions of the Decision Framework
239
11.2 Limitations of Uses of Migration and Population Predictions
241
11.2.1 Alternatives to the Use of Optimal Forecasts
241
11.2.2 Which Questions Can the Forecasts Answer?
244
11.2.3 Towards Interactive Demographic Forecasting?
245
Part V Conclusion
248
12 Summary and Conclusion: Beyond Migration Forecasting
249
12.1 Summary of the Key Findings
249
12.1.1 Bayesian Model Selection and Forecast Averaging
249
12.1.2 Vector Autoregression Models and Their Reduction
250
12.1.3 Models Acknowledging Discontinuity in Trends
251
12.1.4 Sensitivity of the Results to Changes in Priors
252
12.1.5 Ex-ante and Ex-post Comparison of Forecasts: Implications for Users
252
12.1.6 General Conclusions
254
12.2 Bayesian Forecasts in the Population Forecasting Debates
255
12.2.1 Bayesian Methods in Perspective: Uncertainty, Judgement and Occam's Razor
255
12.2.2 Migration Forecasting as a Continuous Process
256
12.2.3 From Point Predictions to Decision Support: In Need of a Paradigm Shift?
259
12.3 A Possible Future of Migration and Its Forecasts
260
Annex A Empirical Illustrations: Data Sources and Preparation
263
Migration Flows
263
Population Stocks
263
Economic Variables
267
Annex B WinBUGS Code Used in the Forecasting Examples
269
Annex C Selected Results of Presented Migration Forecasts
275
References
286
Author Index
305
Subject Index
311