Pilar Rey del Castillo

Survey Methodology

Pilar Rey del Castillo’s research interests focus on survey methodology, specifically on methodologies to produce indicators of evolution over time of phenomena interesting for official statistics. Through the years, she has worked in a variety of projects covering different aspects of survey methodology, from the first steps to find appropriate measures for the observable facts to the last steps of editing and analysing the results. Some specific aspects of interest are: construction of quantitative indicators from business surveys, construction of qualitative indicators from opinion surveys, selective editing methods for quantitative survey data based on time series models, use of data mining and artificial intelligence methods for editing and imputing quantitative and qualitative survey data, use of fuzzy neural networks to impute missing data in surveys, use of Bayesian Networks to impute missing data in surveys, use of Big Data sources to produce new or complementing statistical indicators and development of methods for checking the quality of Big Data for statistical purposes.

Ms Rey del Castillo is currently finishing a paper that relates tax morale with the payment of personal income tax in Spain using a procedure of statistical matching. The selected method to execute the match − imputations using Bayesian Networks − has proved to be particularly suitable because it has two important features: it preserves the distributions of the variables in the original and matched files, and it allows for producing simulations that may be used in subsequent inferences.

While at CESifo, Ms Rey del Castillo plans to access the extensive data available at the Institute and wll try to perform a similar tax-morale study for other countries. She also welcomes participation or collaboration in other studies related to her areas of interest.

Pilar Rey del Castillo is Research Adviser at the Instituto de Estudios Fiscales, Madrid. She has been a member of the UNECE Big Data task team, associate professor at Carlos III University and a statistician at EUROSTAT. She is a graduate in Mathematical Science from Universidad Autónoma de Madrid, holds an MA on Econometric Modelling and Time Series Analysis from Centro de Estudios Monetarios y Financieros, Banco de Espana and a PhD in Computer Science and Artificial Intelligence from the Facultad Informática of the Universidad Politécnica de Madrid.