Spatial Microsimulation with R

Robin Lovelace

• Publisher: Chapman and Hall/CRC
• Year: 2017
• ISBN:
  9781315360669
• Edition: 1
• Topic: Computer Science
• Subtopic: Programming Languages
• Index: Computer Science

Generate and Analyze Multi-Level Data Spatial microsimulation involves the generation, analysis, and modeling of individual-level data allocated to geographical zones. Spatial Microsimulation with R is the first practical book to illustrate this approach in a modern statistical programming language.

Get Insight into Complex Behaviors The book progresses from the principles underlying population synthesis toward more complex issues such as household allocation and using the results of spatial microsimulation for agent-based modeling. This equips you with the skills needed to apply the techniques to real-world situations. The book demonstrates methods for population synthesis by combining individual and geographically aggregated datasets using the recent R packages ipfp and mipfp. This approach represents the "best of both worlds" in terms of spatial resolution and person-level detail, overcoming issues of data confidentiality and reproducibility.

Implement the Methods on Your Own Data Full of reproducible examples using code and data, the book is suitable for students and applied researchers in health, economics, transport, geography, and other fields that require individual-level data allocated to small geographic zones. By explaining how to use tools for modeling phenomena that vary over space, the book enhances your knowledge of complex systems and empowers you to provide evidence-based policy guidance.

Part I

Introducing spatial microsimulation with R

1
Introduction

CONTENTS

1.1 Who this book is for and how to use it

1.2 Motivations

1.3 A definition of spatial microsimulation

1.4 Learning by doing

1.5 Why spatial microsimulation with R?

1.6 Learning the R language

1.7 Typographic conventions

1.8 An overview of the book

1.1 Who this book is for and how to use it

This book is for anyone who wants to learn how to do spatial microsimulation. By that we mean taking individual level and geographical datasets and processing them to solve real-world problems.

This book is a general-purpose introduction to the field and its implementation in a modern programming language. As such there is no single target audience, although the book was written with the following people in mind:

• Practitioners: because of the applied nature of the method and its utility for modelling people, it is useful in a wide range of sectors. Planning for the future of educational, health and transport services at the local level, for example, could benefit from spatial microdata on the inhabitants of the study area. The future is of course uncertain and spatial microsimulation can provide tools for scenario-based planning to explore various visions of the future. In this context spatial microsimulation can be especially useful for designing local policies to assist with the global transition away from fossil fuels (Lovelace and Philips, 2014).
• Academics: spatial microsimulation originated in academic research, building on the work of early pioneers such as Deming and Stephan (1940) and Clarke and Holm (1987). With recent advances in computer power, open source software and data accessibility, spatial microsimulation can now answer a greater range of questions than ever before. Because spatial microsimulation and associated methods are still in their infancy, there are still many areas of research where the approach has yet to be used. This makes spatial microsimulation an ideal method for doing new research, for example as part of a PhD.
• Educators: although spatial microsimulation has become more accessible over the last few years, it is still out of reach for the majority of people. Yet the potential benefits are large. This book provides example data and code that can be used as part of a course involving the analysis of spatial microdata, for example in a module contributing to an undergraduate or MSc course on Transport Modelling, Spatial Economics or Quantitative Geography.

The book has a definite progression from easier content to more advanced topics, especially those in Chapters 11 and 12, which deal with spatial microsimulation for transport modelling and agent-based modelling (ABM), respectively. The book can be read in order from front to back for a detailed overview of the field, its applications and its concepts (Part I); the practicalities and software decisions involved in generating spatial microdata (Part II); and issues around the modelling of spatial microdata (Part III).

Equally, the book can be used as a reference volume, to be dipped into as and when particular topics arise. There is a path dependency in the book however: Part II assumes that the reader is familiar with the concepts introduced in Part I, and the two chapters in Part III assume that the reader is competent with generating and handling spatial microdata, as introduced in Chapters 4 and 5. These are the central chapters in the book, in terms of generating spatial microdata with R. Chapters 4 and 5 are also the most important for people who simply want to generate spatial microdata rapidly. Chapter 6 provides insight into more experimental approaches for generating spatial microdata, while Chapter 7 provides a comprehensive worked example of the methods used ‘in the wild’. A more detailed overview is provided in Section 1.8.

If you can’t wait to get your hands dirty with code and example data for doing spatial microsimulation, please skip to Chapter 3. For other readers who want a little more background on this book and spatial microsimulation, bear with us. The remainder of this chapter explains the thinking behind the book, including the reasons for focussing on spatial microsimulation in R rather than in another language and the motivations for doing spatial microsimulation in the first place. The next chapter is practical. Chapter 3, by contrast, brings the reader up-to-date on what spatial microsimulation is and how it is currently used.

For the structure, each chapter begins with an introduction and ends with a summary of what has been done in the chapter. The last section of this chapter contains an overview of the book.

1.2 Motivations

Imagine a world in which data on companies, households and governments were widely available. Imagine, further, that researchers and decision-makers acting in the public interest had tools enabling them to test and model such data to explore different scenarios of the future. People would be able to make more informed decisions, based on the best available evidence. In this technocratic dreamland pressing problems such as climate change, inequality and poor human health could be solved.

These are the types of real-world issues that we hope the methods in this book will help to address. Spatial microsimulation can provide new insights into complex problems and, ultimately, lead to better decision-making. By shedding new light on existing information, the methods can help shift decision-making processes away from ideological bias and towards evidence-based policy.

The ‘open data’ movement has made many datasets more widely available. However, the dream sketched in the opening paragraph is still far from reality. Researchers typically must work with data that is incomplete or inaccessible. Available datasets often lack the spatial or temporal resolution required to understand complex processes. Publicly available datasets frequently miss key attributes, such as income. Even when high quality data is made available, it can be very diffcult for others to check or reproduce results based on them. Strict conditions inhibiting data access and use are aimed at protecting citizen privacy but can also serve to block democratic and enlightened decision making.

The empowering potential of new information is encapsulated in the saying that ‘knowledge is power’. This helps explain why methods such as spatial microsimulation, that help represent the full complexity of reality, are in high demand.

Spatial microsimulation is a growing approach to studying complex issues in the social sciences. It has been used extensively in fields as diverse as transport, health and education (see Chapter 3), and many more applications are possible. Fundamental to the approach are approximations of individual level data at high spatial resolution: people allocated to places. This spatial microdata, in one form or another, provides the basis for all spatial microsimulation research.

The purpose of this book is to teach methods for doing (not reading about!) spatial microsimulation. This involves techniques for generating and analysing spatial microdata to get the ‘best of both worlds’ from real individual and geographically-aggregated data. Population synthesis is therefore a key stage in spatial microsimulation: generally real spatial microdata are unavailable due to concerns over data privacy. Typically, synthetic spatial microdatasets are generated by combining aggregated outputs from Census results with individual level data (with little or no geographical information) from surveys that are representative of the population of interest.

The resulting spatial microdata are useful in many situations where individual level and geographically specific processes are in operation. Spatial microsimulation enables modelling and analysis on multiple levels. Spatial microsimulation also overlaps with (and provides useful initial conditions for) agent-based models (see Chapter 12).

Despite its utility, spatial microsimulation is little known outside the fields of human geography and regional science. The methods taught in this book have the potential to be useful in a wide range of applications. Spatial microsimulation has great potential to be applied to new areas for informing public policy. Work of great potential social benefit is already being done using spatial microsimulation in housing, transport and sustainable urban planning. Detailed modelling will clearly be of use for planning for a post-carbon future, one in which we stop burning fossil fuels.

For these reasons there is growing interest in spatial microsimulation. This is due largely to its practical utility in an era of ‘evidence-based policy’ but is also driven by changes in the wider research environment inside and outside of academia. Continued improvements in computers, software and data availability mean the methods are more accessible than ever. It is now possible to simulate the populations of small administrative areas at the individual level almost anywhere in the world. This opens new possibilities for a range of applications, not least policy evaluation.

Still, the meaning of spatial microsimulation is ambiguous for many. This book also aims to clarify what the method entails in practice. Ambiguity surrounding the term seems to arise partly because the methods are inherently complex, operating at multiple levels, and partly due to researchers themselves. Some uses of the term ‘spatial microsimulation’ in the academic literature are unclear as to its meaning; there is much inconsistency about what it means. Worse is work that treats spatial microsimulation as a magical black box that just ‘works’ without any need to describe, or more importantly make reproducible, the methods underlying the black box. This book is therefore also about demystifying spatial microsimulation.

1.3 A definition of spatial microsimulation

At this early stage, it is worth considering how spatial microsimulation has been interpreted in past work and how we define the term for the book. Depending on the available data, the aim of the research, and the interpretation of the researcher using the techniques, spatial microsimulation has two broad meanings. Spatial microsimulation can be understood either as a technique or an approach:

1. A method for generating spatial microdata — individuals allocated to zones (see Figure 1.1) — by combining individual and geographically aggregated datasets. In this interpretation, ‘spatial microsimulation’ is roughly synonymous with ‘population synthesis’.
2. An approach to understanding multi level ph...

Table of contents

• Cover Page
• Title
• Copyright
• Contents
• Preface
• Acknowledgements
• List of Figures
• List of Tables
• Part I Introducing spatial microsimulation with R
• Part II Generating spatial microdata
• Part III Modelling spatial microdata
• Glossary
• Bibliography
• Index