Model selection, estimation and forecasting in VAR models with short-run and long-run restrictions

• Published in: Journal of econometrics Vol. 164; no. 1; pp. 116 - 129
• Main Authors: Athanasopoulos, George, de Carvalho Guillén, Osmani Teixeira, Issler, João Victor, Vahid, Farshid
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Amsterdam Elsevier B.V 01.09.2011; Elsevier; Elsevier Sequoia S.A
• Series: Journal of Econometrics
• Subjects: Accuracy; Applications; Cointegration; Distribution theory; Econometric models; Econometrics; Economic forecasts; Exact sciences and technology; Forecasting accuracy; Forecasting techniques; Inference from stochastic processes; time series analysis; Insurance, economics, finance; Mathematics; Model selection criteria; Monte Carlo simulation; Numerical analysis; Numerical analysis. Scientific computation; Numerical methods in probability and statistics; Probability and statistics; Reduced rank models; Reduced rank models Model selection criteria Forecasting accuracy; Regression analysis; Sciences and techniques of general use; Statistics; Studies; Time series; Unit root; Vector-autoregressive models; C53; Model selection criteria; C32; Reduced rank models; Forecasting accuracy; Monte Carlo method; Data analysis; Statistical distribution; Model selection; Finite sample; Prediction theory; Statistical estimation; Autoregressive model; Stochastic method; Stochastic process; Penalty method; Statistical method; Forecasting theory; Numerical analysis; Selection method; Approximation theory; Filtering theory; Econometrics
• ISSN: 0304-4076; 1872-6895
• DOI: 10.1016/j.jeconom.2011.02.009

We study the joint determination of the lag length, the dimension of the cointegrating space and the rank of the matrix of short-run parameters of a vector autoregressive (VAR) model using model selection criteria. We suggest a new two-step model selection procedure which is a hybrid of traditional criteria and criteria with data-dependant penalties and we prove its consistency. A Monte Carlo study explores the finite sample performance of this procedure and evaluates the forecasting accuracy of models selected by this procedure. Two empirical applications confirm the usefulness of the model selection procedure proposed here for forecasting.