Anna Klabunde

Decision-Making in Agent-Based Models of Migration: State of the Art and Challenges.

We review agent-based models (ABM) of human migration with respect to their decision-making rules. The most prominent behavioural theories used as decision rules are the random utility theory, as implemented in the discrete choice model, and the theory of planned behaviour. We identify the critical choices that must be made in developing an ABM, namely the modelling of decision processes and social networks. We also discuss two challenges that hamper the widespread use of ABM in the study of migration and, more broadly, demography and the social sciences: (a) the choice and the operationalisation of a behavioural theory (decision-making and social interaction) and (b) the selection of empirical evidence to validate the model. We offer advice on how these challenges might be overcome.

Modelling and simulating decision processes of linked lives: An approach based on concurrent processes and stochastic race

Individuals’ decision processes play a central role in understanding modern migration phenomena and other demographic processes. Their integration into agent-based computational demography depends largely on suitable support by a modelling language. We are developing the Modelling Language for Linked Lives (ML3) to describe the diverse decision processes of linked lives succinctly in continuous time. The context of individuals is modelled by networks the individual is part of, such as family ties and other social networks. Central concepts, such as behaviour conditional on agent attributes, age-dependent behaviour, and stochastic waiting times, are tightly integrated in the language. Thereby, alternative decisions are modelled by concurrent processes that compete by stochastic race. Using a migration model, we demonstrate how this allows for compact description of complex decisions, here based on the Theory of Planned Behaviour. We describe the challenges for the simulation algorithm posed by stochastic race between multiple concurrent complex decisions.

Computational Economic Modeling of Migration

In this paper an agent-based model of endogenously evolving migrant networks is developed to identify the determinants of migration and return decisions. Individuals are connected by links, the strength of which declines over time and distance. Methodologically, this paper combines parameterization using data from the Mexican Migration Project with calibration. It is shown that expected earnings, an idiosyncratic home bias, network ties to other migrants, strength of links to the home country and age have a significant impact on circular migration patterns. The model can reproduce spatial patterns of migration as well as the distribution of number of trips of migrants. It is shown how it can also be used for computational experiments and policy analysis.

Multistate modelling extended by behavioural rules: An application to migration

We propose to extend demographic multistate models by adding a behavioural element: behavioural rules explain intentions and thus transitions. Our framework is inspired by the Theory of Planned Behaviour. We exemplify our approach with a model of migration from Senegal to France. Model parameters are determined using empirical data where available. Parameters for which no empirical correspondence exists are determined by calibration. Age- and period-specific migration rates are used for model validation. Our approach adds to the toolkit of demographic projection by allowing for shocks and social influence, which alter behaviour in non-linear ways, while sticking to the general framework of multistate modelling. Our simulations yield that higher income growth in Senegal leads to higher emigration rates in the medium term, while a decrease in fertility yields lower emigration rates.

Trust in a Network of Investors and Startup Entrepreneurs

We propose a simulation model of a stylized angel investor market in which the business relations are conditioned on the trust of the angel into the startup entrepreneur. Initial trust depends on social proximity between the angel and the entrepreneur and is later on updated depending on the returns an investor receives from an entrepreneur. We show that investors benefit most in terms of returns from an intermediate sensitivity of trust to returns. From an investor’s perspective, too much trust keeps too many unproductive firms in the market, while too little trust leads to the termination of profitable relations because of minor productivity drops. The proportion of entrepreneurs who reach the objective of leaving the market with enough capital, however, is higher if investors lose trust quickly and stop funding early. In this sense there is a conflict of interest between investors and entrepreneurs.

The science of choice: an introduction

Introduction In October 2015, around 30 scholars convened at the Max Planck Institute for Demographic Research (MPIDR) in Rostock to discuss: (a) how individuals and families make decisions about marriage, child-birth, migration, retirement, and other transitions in the life course; and (b) how these decision processes can be operationalized in demographic models. The workshop was organized by the Scientific Panel on Microsimulation and Agent-Based Modelling con- vened by the International Union for the Scientific Study of Population (IUSSP) and by MPIDR. The report of this ‘Science of choice’ workshop and the papers presented are available from the workshop’s website (see IUSSP 2015). The five papers included in this Supplement are revised versions of papers presented at the workshop in Rostock.

ML3: a language for compact modeling of linked lives in computational demography

Agent-based modeling and simulation is widely used in computational demography. Although existing agent-based approaches allow modeling linked lives in a rather flexible manner, the resulting models, due to typically being implemented in a general-purpose programming language, often lack the compactness required to easily access the model. With ML3 (Modeling Language for Linked Lives) we present a compact and expressive domain-specific modeling language for continuous-time agent-based models in computational demography. The language combines elements from guarded commands, process algebras, and rule-based approaches. We discuss and present the individual features of the language and illuminate its compactness by presenting the specification of an entire agent-based model from recent literature.

How to Describe Agent-Based Models in Population Studies?

Agent-based models (ABMs) are increasingly used to describe, understand and predict human behaviour. The flexibility of agent-based models allows exploring mechanisms and feedback effects. However, this flexibility comes at a price, which is the problem of communicating and describing the model in a concise and transparent manner. There is no general agreement or culture yet on how agent-based models should be documented and described. Therefore, we want to contribute to the discussion on how agent-based models should be described in this position paper from a demographic perspective. Firstly we present and discuss the plethora of different ways to describe published agent-based models in population studies and demography. Subsequently, we argue that the ODD + D (Overview, Design, Details + Decisions) protocol is apt to address the specific difficulties which arise in describing demographic ABMs. We describe the main elements of the ODD + D protocol and subsequently present the results of an experiment illustrating its usefulness to describe an agent-based model of circular migration. We conclude that protocols such as the ODD + D are not only useful to compile a model description, but also serve as tool for the model developer to check if she has provided all necessary information that allows the reader to fully understand and replicate the model.