Agustín Maravall

After a childhood in Paris, Maravall returned to Madrid and finished secondary school at the Colegio Estudio, completed a doctorate in agricultural engineering at the Technical University of Madrid, and worked for several years at the Spanish Ministry of Agriculture. With a Fulbright-Ford fellowship he moved to the US and completed a Ph.D. in Economics at the University of Wisconsin-Madison. In 1975, he moved to Washington, D.C. to work as staff economist in the Research Division of the Federal Reserve Board of Governors. In 1979, he returned to Madrid as a senior economist in the Research Department of the Bank of Spain, and in 1989 he moved to Italy as Full Professor in the Department of Economics of the European University Institute in Florence. In 1996, he returned to the Bank of Spain as Chief Economist and Head of the Time Series Analysis Unit.

Maravall has been on the editorial board of many professional journals (the lengthiest periods for the Journal of Business and Economic Statistic and the Journal of Econometrics), on the Program and Scientific Committees of many international meetings and conferences, and has taught courses in more than 30 countries to participants from more than 60 countries. He has been Special Advisor to the European Central Bank and of Eurostat, and a member of the Board of Directors of the former Institute for Advanced Studies in the Social Sciences of Madrid and of the High Advisory Council for R & D of the Generalitat Valenciana.

Maravall’s research has centered on time series analysis and modeling and their application to economic series. His main contribution (an important part of it coauthored with Victor Gómez) has been the development of a model-based procedure to jointly solve several statistical time series problems that affect analysis and interpretation of economic time series. The standard (default) procedure performs, first, automatic identification and forecasting of regression-ARIMA models (that includes adjusting for outliers and calendar effects) in the possible presence of missing observations.[1] Then, the model is decomposed into models for the unobserved components (such as seasonal, trend, transitory, and cyclical components) and, from these models, filters are derived to estimate and forecast the components.[2][3][4] The model-based structure provides the joint distribution of the estimators, from which parametric tests and inferences (such as the standard errors of all estimators and forecasts) are derived.[5][6]

Relevant features of the procedure are, first, that—for a given series—all estimators are derived from the same (well-defined) model and hence will be internally consistent. Second, the automatic performance can be efficient and reliably applied to very large sets of time series.[7][8]

The research has been published in academic journals (with some reprinted in books of readings), in the Springer-Verlag monograph series Lecture Notes in Statistics and Lecture Notes in Economics and Mathematical Systems, in the Bank of Spain monograph series, and in many book chapters.[9]

With the collaboration of Víctor Gómez (1987–1999) and Gianluca Caporello (1990–2015), the research was incorporated into several computer programs: TRAMO (“Time series Regression with ARIMA noise, Missing values, and Outliers”), SEATS (“Signal Extraction in ARIMA Time Series”), and TSW (“Tramo and Seats for Windows”). An extension of TRAMO, TERROR (“TRAMO for errors”), contains an application to data editing (it detects possible errors in incoming data, in large time-series data sets).

The programs are extensively used throughout the world in a variety of applications. An important one is official production of seasonally adjusted data, and their use has been recommended by working groups at many institutions.[10] The recent X13-ARIMA-SEATS program of the US Bureau offers two options: one is their program X12-ARIMA (that had adopted and adapted the automatic model identification procedure of TRAMO); the other option is SEATS.[11] The European Statistical System (central banks plus statistical agencies) has developed JDEMETRA+, basically a JAVA interface with X12-ARIMA and TRAMO-SEATS that is recommended for European countries.[12] TRAMO and SEATS are also included in statistics and econometrics packages, and are freely available from the Bank of Spain.[13]

Former Wisconsin Alumni Research Foundation Fellow (1973–75); Fulbright-Ford Fellow (1971–73); Phi Kappa Phi Honor Society Fellow (1975); Oficial de la Orden Civil del Mérito Agrario (1971). Elected member of the International Statistical Institute (1988).

Fellow of the Journal of Econometrics, 1995; Fellow of the American Statistical Association, 2000.

2004 Julius Shiskin Award for Economic Statistics, sponsored by the Washington Statistical Society, the National Association for Business Economics, and the American Statistical Association, Business and Economics Section. Rey Jaime I Prize in Economics, 2005, sponsored by the Spanish Royal House, the "Fundación Premios Rey Jaime I,” and the Generalitat Valenciana. First Galicia Prize in Statistics, 2006, sponsored by the Caixa Galicia Foundation and the Galicia Statistical Institute. Rey Juan Carlos Prize in Economics, 2014, sponsored by the Celma Foundation and presented by the King of Spain.

Homage “Celebrating 25 years of TRAMO-SEATS and the 70th birthday of Agustín Maravall,” hosted by the Bank of Spain, March 2014, with the participation of researchers from universities, central banks and statistical agencies in Belgium, France, Germany, Italy, The Netherlands, Spain, Sweden, the UK, and the USA.[14]