Foreword 3

Executive summary 8

1 Introduction 10

For whom is this handbook? 11

Forecasting for migration flows by categories versus (net) migration as a whole? 12

References 13

2 Current state of play 14

Where do OECD countries stand on migration forecasting? 14

Notes 15

3 What is the question to which forecasting is expected to respond? 16

What is the purpose of migration forecasting, and what are the associated challenges? What are policymakers looking for through forecasting? 16

References 19

4 Forecasters’ toolkit: Choosing the right model for the task 20

What are the existing forecasting methods and models? 20

What is the appropriate forecasting model for each migration category? 24

What resources do countries need to run forecasting? 29

How complex should the model be? 31

How often should model results be updated? 32

References 33

5 What data are necessary to conduct the forecasting exercise? 37

How to assess the existing sources of data? 37

Where to find and how to incorporate qualitative data? 41

How to quantify policy indicators for statistical models? 44

Which software is the best fits for forecasting, causal effect estimation and projection? 47

References 49

Notes 52

6 How should the forecasting model be developed? 53

Is it still accurate to use traditional statistical and econometric time series analysis? 53

What are Bayesian Models, their principles, and key features? 55

How to include contextual knowledge in forecasting: Expert opinion and Delphi surveys 56

How to run a Delphi survey and how to include expert replies in the model? 57

How to develop machine learning models? 60

How to incorporate migration drivers into the machine learning models? 61

How to model causal influence in migration forecasting? 62

References 64

7 How can the model be evaluated? 67

How to validate model performance? 67

How to calibrate the model? 69

How to assess robustness? 71

How to back-test ML/Stat models? 72

How can expert knowledge and scenarios be back-tested? 72

References 74

8 How should forecasts be presented to policymakers and ensure effective interaction between policymaking and forecasting? 76

How to present forecasts and their inherent uncertainty to policymakers 76

How to use the forecast properly and effectively 82

References 84

Notes 85

9 What future developments can be expected in migration forecasting? 86

How to measure the potential effect of policies in countries of destination? 86

Is conformal prediction the final frontier in time series forecasting? 87

How can forecasting results in one migration category affect forecasting in other categories? 88

How can forecasts in one country affect forecasts in other countries (i.e. interaction across countries)? 89

References 90

Annex A. List of Migration Anticipation and Preparedness (MAP) task force members 92

Glossary 93