98 research outputs found
Econometric modelling of time series with outlying observations
No full textEconomies are buffeted by natural shocks, wars, policy changes, and other unanticipated events. Observed data can be subject to substantial revisions. Consequently, a “correct” theory can manifest serious mis-specification if just fitted to data ignoring its time-series characteristics. Modelling U.S. expenditure on food, the simplest theory implementation fails to describe the evidence. Embedding that theory in a general framework with dynamics, outliers and structural breaks and using impulse-indicator saturation, the selected model performs well, despite commencing with more variables than observations (see Doornik, 2009b), producing useful robust forecasts. Although this illustration involves a simple theory, the implications are generic and apply to sophisticated theorie
Understanding economic forecasts
No full textNine articles, originally presented at the Annual Festival of Science at the University of Sheffield in September 1999, explain new developments in economic forecasting. Papers examine how economists forecast (David F. Hendry); economic modeling for fun and profit (Paul Turner); making sense of published economic forecasts (Diane Coyle); forecast uncertainty in economic modeling (Neil R. Ericsson); evaluation of forecasts (Clive W. J. Granger); forecasting and the UK business cycle (Denise R. Osborn, Marianne Sensier, and Paul W. Simpson); modeling and forecasting at the Bank of England (Neal Hatch); forecasting the world economy (Ray Barrell); and the costs of forecast errors (Terence Burns). Hendry is Professor of Economics at Nuffield College, Oxford University. Ericsson is a staff economist at the Division of International Finance, Federal Reserve Board. Author and subject indexes
Understanding economic forecasts
No full textNine articles, originally presented at the Annual Festival of Science at the University of Sheffield in September 1999, explain new developments in economic forecasting. Papers examine how economists forecast (David F. Hendry); economic modeling for fun and profit (Paul Turner); making sense of published economic forecasts (Diane Coyle); forecast uncertainty in economic modeling (Neil R. Ericsson); evaluation of forecasts (Clive W. J. Granger); forecasting and the UK business cycle (Denise R. Osborn, Marianne Sensier, and Paul W. Simpson); modeling and forecasting at the Bank of England (Neal Hatch); forecasting the world economy (Ray Barrell); and the costs of forecast errors (Terence Burns). Hendry is Professor of Economics at Nuffield College, Oxford University. Ericsson is a staff economist at the Division of International Finance, Federal Reserve Board. Author and subject indexes
Encompassing
No full textWith the kind permission of the Editors of the Oxford Bulletin, this Special Issue onEncompassing is a volume in honour of Grayham E. Mizon, to celebrate his majorcontributions to econometrics, and in particular, to the development of the theory ofencompassing and testing non-nested hypotheses. Grayham’s publications on encom-passing by himself, and with various co-authors (including most contributors to thisSpecial Issue), have been cited more than 800 times, reflecting its widespread useacross a diverse range of empirical and theoretical studies.Grayham has advanced numerous other areas in econometrics, including modelselection and progressive research strategies, panel data and time-series analyses,simulation and Monte Carlo methods, forecast evaluation and economic policy anal-ysis, and exogeneity, as well as encompassing. In addition to the formulation andimplementation of new estimators and test statistics, he has contributed to the sub-stantive application of new econometric tools, developing empirical models andpolicy analyses for both developed and transition economies, investigating consump-tion, employment and output, wage and price inflation, and relative prices. A partialbibliography is provided below.Grayham was also a prolific ‘producer’ of doctoral students – several of whomhave contributed to this Special Issue – as well as a constructive referee and editor,both for specific projects and in general for the highly successful Advanced Texts inEconometrics, which he overviewed with Clive W.J. Granger on behalf of OxfordUniversity Press.We are delighted to have been able to collect together a set of papers whichfurther advance the coverage of encompassing, by investigating the principles ofencompassing; encompassing as a key part of a general empirical methodology;its role in both automatic and human-driven model selection; hypothesis testingfor linear and nonlinear, nested and non-nested, parametric and non-parametricmodels; as well as simulation, forecast, cross-data vintage, and Bayesianencompassing.Until he reads this Foreword, Grayham will have been unaware of this purposeof our Special Issue, despite being the main Guest Editor, and co-author of two ofthe papers. We are grateful to all the contributors for sustaining our secret, and hopeit is a pleasant surprise for Grayham. Most authors and co-authors have been able to offer their own felicitations toGrayham in another paper – one which was not written with him – except for ChiaraMonfardini, who therefore joins the two other Guest Editors in congratulating Gray-ham on a truly fecund career, and wishing him continuing success in unravelling thetheory and practice of econometrics.-- Foreword, David F. Hendry, Massimiliano Marcellino and Chiara Monfardini
-- Guest Editors’ Introduction to Special Issue on Encompassing, David F. Hendry, Massimiliano Marcellino and Grayham E. Mizon
-- Encompassing: Concepts and Implementation, Christophe Bontemps and Grayham E. Mizon
-- Encompassing Procedures, Christophe Bontemps, Jean-Pierre Florens and Jean-François Richard
-- Simulation Encompassing: Testing Non-nested Hypotheses, Maozu Lu, Grayham E. Mizon and Chiara Monfardini
-- Log Income vs. Linear Income: An Application of the Encompassing Principle, Luigi Ermini and David F. Hendry
-- Linear vs. Log-linear Unit-Root Specification: An Application of Mis-specification Encompassing, Aris Spanos, David F. Hendry and J. James Reade
-- Cross-data-vintage Encompassing (pages 849–865) Steve Cook
-- Model Selection for Nested and Overlapping Nonlinear, Dynamic and Possibly Mis-specified Models, Massimiliano Marcellino and Barbara Rossi
-- The Fragility of Sensitivity Analysis: An Encompassing Perspective, Neil R. Ericsson
-- Encompassing and Automatic Model Selection, Jurgen A. Doornik
-- An Omnibus Test for Univariate and Multivariate Normality, Jurgen A. Doornik and Henrik Hanse
Econometric Modeling: A Likelihood Approach
No full textPresents a likelihood-based introduction to econometrics. Discusses the Bernoulli model; inference in the Bernoulli model; a first regression model; the logit model; the two-variable regression model; the matrix algebra of two-variable regression; the multiple regression model; the matrix algebra of multiple regression; misspecification analysis in cross sections; strong exogeneity; empirical models and modeling; autoregressions and stationarity; misspecification analysis in time series; the vector autoregressive model; identification of structural models; nonstationary time series; cointegration; Monte Carlo simulation experiments; automatic model selection; structural breaks; forecasting; and the way ahead. Hendry is Professor of Economics at the University of Oxford and a Fellow of Nuffield College. Nielsen is Reader in Econometrics at the University of Oxford and a Fellow of Nuffield College. Author and subject indexes
Recommended from our members
Forecasting economic time series
No full textAnalyzes the models, procedures, and measures of economic forecasting with a view to improving forecasting practices. This volume sets the scene by focusing on forecasting when the underlying process can be described by a stationary representation. A companion volume, Zeuthen Lectures on Economic Forecasting will discuss forecasting in the presence of deterministic nonstationarities. Provides an introduction to forecasting. Discusses first principles; forecasting in univariate processes; Monte Carlo simulation techniques; forecasting in cointegrated systems; forecasting with large-scale macroeconometric models; the value of intercept corrections; forecasting using leading indicators; combining forecasts; multistep estimation; the value of imposing parsimony; and testing forecast accuracy. Sets out recommendations for forecasting practice. Clements is Research Fellow in Economics at the University of Warwick. Hendry is Leverhulme Personal Research Professor of Economics and Fellow of Nuffield College, Oxford University. Author and subject indexes
Econometrics: Alchemy or science? Essays in econometric methodology
No full textEighteen previously published papers discuss econometric methodology. Essays focus on econometrics--alchemy or science; stochastic specification in an aggregate demand model of the United Kingdom; dynamic specification; testing dynamic specification in small simultaneous systems; the times-series approach to econometric model building; serial correlation as a convenient simplification; an empirical application and Monte Carlo analysis of tests of dynamic specification; econometric modeling of the aggregate time-series relationship between consumers' expenditure and income in the United Kingdom; liquidity and inflation effects on consumers' expenditure; interpreting econometric evidence and the behavior of consumers' expenditure in the United Kingdom; predictive failure and econometric modeling in macroeconomics--the transactions demand for money; monetary economic myth and econometric reality; the structure of simultaneous equations estimators; AUTOREG--a computer program library for dynamic econometric models with autoregressive errors; exogeneity; the formulation of empirical models in dynamic econometrics; the econometric analysis of economic time series; and econometric modeling and the consumption function in retrospect. Also includes a postscript on the econometrics of PC-GIVE. Preambles to each essay sketch the key points; the lessons learned by the author and the developments triggered; and the crucial issues missed at the time. Hendry is Professor of Economics at the University of Oxford and Fellow of Nuffield College. Bibliography; index
Dynamic econometrics
No full textProvides tools for the critical appraisal of empirical evidence in time-series econometrics as well as an organized approach to econometrics that will enable the reader to undertake applied econometrics research and establish a credible evidential basis for the results. Designed for economists investigating empirical phenomena, for advanced undergraduate and graduate econometrics students, and for statisticians involved in the analysis of social science time series. Part 1 addresses the situation where the complete structure of the process that generates economic data is known and the values of its parameters, discussing important econometric concepts, econometric tools and techniques, dynamics and interdependence, exogeneity and causality, interpretations of linear models, a typology of linear dynamic equations, dynamic systems, and the theory of reduction. Part 2 studies estimation and inference, assuming the form of the data-generation process is known but that its parameters are unknown, and covers likelihood, simultaneous equations systems, measurement problems in econometrics, and testing and evaluation. Part 3 addresses practical problems, modeling, and applications in the empirically relevant setting where even the form of the data-generation process is unknown to the investigator, including issues of model discovery, evaluation, data mining, model misspecification, and encompassing. Includes end-of-chapter exercises. Hendry is Professor of Economics at Oxford University and a fellow of Nuffield College. Author and subject indexes
Recommended from our members
Forecasting non-stationary economic time series
No full textAddresses the problems confronting forecasting in economies subject to structural breaks. Discusses economic forecasting and provides a taxonomy of the sources of forecast failure. Explores the role of unmodeled shifts in parameters as a source of misprediction and considers other potential sources of forecast failure, such as model misspecification, parameter estimation uncertainty, a lack of parsimony in model selection, and forecast origin mismeasurement. Analyzes the likely effects of each source of forecast failure via Monte Carlo simulation and empirical examples. Investigates analytically whether time-series models in differences can outperform econometric models in forecasting when a structural break has occurred. Evaluates whether intercept corrections help to robustify forecasts in the face of change. Illustrates the effects of structural breaks and the performance of the various forecasting strategies in a model of consumer expenditure in the United Kingdom and in a model of U.K. money demand. Addresses the removal of deterministic shifts in systems of equations using linear combinations of variables. Considers the forecasting performance of some regime-switching autoregressive models relative to linear autoregressive alternatives. Illustrates a number of the forecasting approaches in a model of wages, prices, and unemployment. Clements is Research Fellow in Economics at the University of Warwick. Hendry is the Leverhulme Personal Research Professor of Economics and Fellow of Nuffield College at Oxford University. Glossary; author and subject indexes
Computationally-intensive Econometrics using a Distributed Matrix-programming Language
No full textThis paper reviews the need for powerful facilities in econometrics, focusing on concrete problems which arise in financial economics and in macroeconomics. We argue that the profession is being held back by the lack of easy to use generic software which is able to exploit the availability of cheap clusters of distributed computers. Our response is to extend, in a number of directions, the well known matrix-programming interpreted language Ox developed by the first author. We note three possible levels of extensions: (i) Ox with parallelization explicit in the Ox code; (ii) Ox with a parallelized run-time library; (iii) Ox with a parallelized interpreter. This paper studies and implements the first case, emphasizing the need for deterministic computing in science. We give examples in the context of financial economics and time-series modelling.Distributed computing; Econometrics; High-performance computing; Matrix-programming language
- …
