Advancing Microdata Models and Methods

This research develops and applies tools to extract information about individual behaviour, and influences on it, from data on individuals' actions and outcomes in their normal environment. Typically, this environment is not a controlled laboratory-like situation. The 'individuals' the project studies can be, for example, people, households or enterprises.
The availability, scope, and scale of microdata are growing very fast, as is the computing technology to process this so-called "big data". Many weighty public sector policy decisions and private sector decisions on pricing, investment and strategy are based on the results of analysis of microdata.
Analysis of microdata presents great challenges that this project will tackle. Some of these arise because the processes in which individuals are engaged are complex, involving risk and uncertainty set in a dynamic context in which the outcomes of decisions occur long after decisions are made.
Saving and pension planning behaviour is of this type. It is critically important for government and firms, for example insurance companies, to understand such behaviours. Further complication arises because there is enormous heterogeneity. People have very different attitudes to risk or make very different estimates of their likely needs or longevity. This heterogeneity may produce big differences in savings outcomes, possibly resulting in great inequality in circumstances amongst the old. Housing location, purchase and rental outcomes are determined in a similarly complex heterogeneous process.
This project will produce new models, methods and approaches for understanding behaviour of this sort, and it will apply them to develop answers to real policy questions using real microdata.
An integral part of the project is a major programme of: hands-on basic training in microdata methods; advanced training given in masterclasses on themes at our research frontier; workshops giving themed presentations of research and stimulating new thinking amongst users and researchers. These activities will be conducted under the auspices of CeMMAP, the ESRC Centre for Microdata Methods and Practice.
The research is organised under five headings.
1) Big data and machine learning. We will study the performance of big data and machine learning methods, develop models and research applications in which economic and social science poses deep questions and guides model construction. Applications will study the impact of policies to change behaviour using taxes on e.g. sugar and fat, and the impact of planning laws on housing supply and prices.
2) Dynamics and Complexity. We will develop models and tools to understand influences on complex life planning choices made by people in the face of uncertainty and incomplete information. Applications will study e.g. housing location and tenure choice.
3) Robust models. We will study the utility and performance of incomplete models of complex economic and social process in which only theory-grounded restrictions on behaviour are employed. Applications will consider e.g. models of auction bidding behaviour, determination of well-being and firm market entry and product choice.
4) Networks and interactions. We will study properties of tools to estimate aspects of network structure and develop methods for estimating the impact of network structure on outcomes. Applications will consider e.g. employer-employee matching processes and trade flows between countries.
5) Survey design and measurement. We will research the optimal choice of scope, scale and measurement quality in investigations such as field experiments constructed to answer specific social science questions.
This work, delivering tools for robust analysis of emerging microdata resources, will be the core of CeMMAP's ongoing research programme, sustaining its outreach engagement and networking activities through to 2021.

This Project will benefit:
(A) Social science researchers, who use microdata to understand human behaviour and the influences on it.
(B) Statisticians, both practitioners and theorists, who develop statistical procedures and research their properties.
(C) Survey designers who gather data on individuals' circumstances, decisions and outcomes, and designers and analysts of policy evaluations and of social and economic lab and field experiments.
(D) Policy and decision makers in the public and private sectors who use microdata evidence to inform policy choices and analysts who process data to inform decisions.
(E) The general public.

(A) and (B) The Project's research will improve the tools used to extract information from microdata directly benefitting these groups. The Project's dissemination and training at basic and advanced levels will spread knowledge of new tools, raising the quality of empirical social science research.
(C) Designers of surveys, evaluations and experiments will benefit from the Project's research on identification, big data and survey design. The results will offer improved guidance on what to measure, how to measure it, and other aspects of design, as well as new tools to analyse big data. Private sector survey designers will be able to use our tools for more efficient data collection, enabling better informed choices on prices, investment and product portfolios. Our work on measurement and big data will serve as an input to improve UK official statistics following the Independent Review of UK Economic Statistics (Bean, 2016).
(D) Policy and decision makers will benefit from the Project's research and activities via better design of policy evaluations, higher-quality analysis of microdata and better understanding of responses to policy changes. Our work on networks and interactions will enable new insights into the workings of education systems, labour markets, and trade and supply chains. Our work on dynamics and complexity will inform thinking about housing choices, savings choices and retirement behaviour. Our work on big data and machine learning methods will unlock huge data resources available to companies and government. Our work on robust models will produce more realistic assessments of the knowledge base on which policy is constructed, resulting in improved decision making and more effective public policy. This will raise the efficiency of government and industry with resulting benefits to customers, shareholders and the population in general.
(E) The general public is often presented with statistics as if they give a direct insight into causal relationships. The Project's work on structural modelling helps separate causation from correlation. Our work on robust models will lead to more cautious and realistic assessments of forecast quality. The research and capacity building of the Project will contribute to raising the sophistication of arguments in public discussions of microdata inferences. Some Project outputs will be published as briefing notes for public consumption.
To ensure impacts are achieved, our research programme is tied in with a systematic user engagement programme consisting of training courses, masterclasses and conferences designed to spread knowledge of new research findings, elicit information from users on pressing research questions and spark new collaborations. Our research and software developed will be published at the CeMMAP web site. As set out in Pathways to Impact, we will continue to develop contacts and collaborate with government and private sector users, a process made easy by our location at IFS.
Benefits from the Project's work will range from immediate (e.g. improved skills from training) to long-term (e.g. accumulation of knowledge about behaviour). Intermediate benefits will also arise as novel methods and approaches deliver higher-quality evidence and better policy choices.