Mark Birkin

The estimation of population microdata by using data from small area statistics and samples of anonymised records

Census data can be represented both as lists and as tabulations of household/individual attributes. List representation of Census data offers greater flexibility, as the exploration of interrelationships between population characteristics is limited only by the quality and scope of the data collected. Unfortunately, the released lists of household/individual attributes (Samples of Anonymised Records, SARs) are spatially referenced only to areas (single or merged districts) with populations of 120 000 or more, whereas released tabulations are available for units as small as single enumeration districts (Small Area Statistics, SAS). Intuitively, it should be possible to derive list-based estimates of enumeration district populations by combining information contained in the SAR and the SAS. In this paper we explore the range of solutions that could be adapted to this problem which, ultimately, is presented as a complex combinatorial optimisation problem. Various techniques of combinatorial optimisation are tested, and preliminary results from the best performing algorithm are evaluated. Through this process, the lack of suitable test statistics for the comparison of observed and expected tabulations of population data is highlighted.

The Generation of Individual and Household Incomes at the Small Area Level using Synthesis

BIRKIN M. and CLARKE M. (1989) The generation ofindividual and household incomes at the small area level using synthesis, Reg. Studies 23, 535–548. This paper describes a method for generating estimates of small area income distributions at the micro-level. The methodology is based on the combination of iterative proportional fitting to estimate missing data and micro-simulation. This method generates synthetic individuals and households together with their associated attributes for enumeration districts in the study area. Data sets that are used include the 1981 Census of Population together with the New Earnings Survey. Results from the modelling exercise are used to illustrate both the types of information that are produced and the flexible nature of aggregation when a micro-data approach is adopted. The study area used is the Leeds Metropolitan District but the results should prove of more general interest. BIRKIN M. et CLARKE M. (1989) La generation des revenus individuels et menagers par petites zon...

A spatial microsimulation model with student agents

In this paper, we present a dynamic simulation model which projects the future population of the city of Leeds as a basis for policy analysis and scenario planning. We argue that microsimulation modelling is not entirely effective in the representation of student populations. Alternative approaches using both spatial interaction models and student agents are presented and evaluated. The results from the agent-based model are found to be particularly encouraging. We suggest that agent-based modelling and microsimulation are powerful as complementary technologies for individual-based modelling.

A Review of Microsimulation and Hybrid Agent-Based Approaches

In this chapter we introduce an approach to individual based modelling of social and economic systems. Microsimulation models (MSM) appear similar to ABM through the representation of individual decision-making units, but there is a significant variation of emphasis between the two approaches. MSM are typically stochastic or rule-based, and with a strong applied policy focus. These characteristics are explored and elaborated through a number of examples. While MSM are often very rich in their representation of ‘structures’, ABM are usually better tuned to the analysis of ‘behaviours’. We therefore argue that there is a strong logic to considering the MSM and ABM approaches as complementary and to begin a search for hybrids which might combine the best features of both approaches.

A Hybrid Multi-Agent/Spatial Interaction Model System for Petrol Price Setting

This paper examines the use of multi-agent simulations (MAS) to model the petrol market. The development of a purely agent based model for petrol prices is presented. Failings within this model are discussed and an alternative strategy for controlling the price of each petrol station based on population of customers is considered. Individual level modelling of customers is too computationally intensive; therefore a spatial interaction model was used to estimate the sales and linked to the agent system to create a hybrid model. To evaluate how effective this hybrid model was, a comparison was made with an existing data set of real petrol prices collected over a two month period. This was achieved both statistically and visually with the use of a Geographical Information System (GIS). Experimentation revealed that the hybrid model outperformed the agent model. Investigation into the behaviour and sensitivity of the system (for example, how prices diffuse spatially) was undertaken by means of idealised simulations.

Genetic Algorithm Optimisation of An Agent-Based Model for Simulating a Retail Market:

Traditionally, researchers have used elaborate regression models to simulate the retail petrol market. Such models are limited in their ability to model individual behaviour and geographical influences. Heppenstall et al presented a novel agent-based framework for modelling individual petrol stations as agents and integrated important additional system behaviour through the use of established methodologies such as spatial interaction models. The parameters for this model were initially determined by the use of real data analysis and experimentation. This paper explores the parameterisation and verification of the model through data analysis and by use of a genetic algorithm (GA). The results show that a GA can be used to produce not just an optimised match, but results that match those derived by expert analysis through rational exploration. This may suggest that despite the apparent nonlinear and complex nature of the system, there are a limited number of optimal or near optimal behaviours given its constraints, and that both user-driven and GA solutions converge on them.

Spatial Microsimulation Models: A Review and a Glimpse into the Future

In this chapter we present a review of the development of microsimulation modelling (MSM) over the past 50 years or so and attempt to outline some of the challenges and opportunities that researchers in the field are currently exploring. Phil Rees is perhaps best known for his research in fields outside MSM but, as we will indicate, he has made significant contributions largely through collaboration and supervision of research students at Leeds, so it is fitting that in this book there is a chapter that makes due acknowledgement of his work.

Elements of a computational infrastructure for social simulation

Applications of simulation modelling in social science domains are varied and increasingly widespread. The effective deployment of simulation models depends on access to diverse datasets, the use of analysis capabilities, the ability to visualize model outcomes and to capture, share and re-use simulations as evidence in research and policy-making. We describe three applications of e-social science that promote social simulation modelling, data management and visualization. An example is outlined in which the three components are brought together in a transport planning context. We discuss opportunities and benefits for the combination of these and other components into an e-infrastructure for social simulation and review recent progress towards the establishment of such an infrastructure.

Assessing the long-term performance of cross-sectoral strategies for national infrastructure

National infrastructure systems (energy, transport, digital communications, water, and waste) provide essential services to society. Although for the most part these systems developed in a piecemeal way, they are now an integrated and highly interdependent “system of systems.” However, understanding the long-term performance trajectory of national infrastructure has proved to be very difficult because of the complexity of these systems (in physical and institutional terms) and because there is little tradition of thinking cross-sectorally about infrastructure system performance. Here, a methodology is proposed for analyzing national multisectoral infrastructure systems performance in the context of uncertain futures, incorporating interdependencies in demand across sectors. Three contrasting strategies are considered for infrastructure provision (capacity intensive, capacity constrained, and decentralized) and multiattribute performance metrics are analyzed in the context of low, medium, and high demographic and economic growth scenarios. The approach is illustrated using Great Britain and provides the basis for the development and testing of long-term strategies for national infrastructure provision. It is especially applicable to mature industrial economics with a large stock of existing infrastructure and challenges of future infrastructure provision.

Analysis of crime patterns through the integration of an agent-based model and a population microsimulation

In recent years, criminologists have become interested in understanding crime variations at progressively finer spatial scales, right down to individual streets or even houses. To model at these fine spatial scales, and to better account for the dynamics of the crime system, agent-based models of crime are emerging. Generally, these have been more successful in representing the behaviour of criminals than their victims. In this paper it is suggested that individual representations of criminal behaviour can be enhanced by combining them with models of the criminal environment which are specified at a similar scale. In the case of burglary this means the identification of individual households as targets. We will show how this can be achieved using the complementary technique of microsimulation. The work is significant because it allows agent-based models of crime to be refined geographically (to allow, for example, individual households with varying wealth or occupancy measures) and leads to the identification of the characteristics of individual victims.