TY - JOUR
T1 - Efficient estimation of the parameter path in unstable time series models
AU - Müller, Ulrich K.
AU - Petalas, Philippe Emmanuel
N1 - Funding Information:
Acknowledgements. We benefitted from thoughtful and constructive comments and suggestions by the editor, Enrique Sentana, and two anonymous referees. We would also like to thank Mark Watson, as well as participants at the NBER Summer Institute, the Workshop for Nonlinear and Nonstationary Models at the California Institute of Technology, the Unit Root and Cointegration Testing Conference in Faro, the Econometric Society World Congress in London, and workshops at the University of Lausanne, New York University, Rutgers University, University of Texas at Austin, FRB of Atlanta, and Iowa State University for useful discussions, and Edouard Schaal for excellent research assistance. Müller gratefully acknowledges financial support from the NSF through grant SES-0518036.
PY - 2010/10
Y1 - 2010/10
N2 - The paper investigates inference in non-linear and non-Gaussian models with moderately time-varying parameters. We show that for many decision problems, the sample information about the parameter path can be summarized by an artificial linear and Gaussian model, at least asymptotically. The approximation allows for computationally convenient path estimators and parameter stability tests. Also, in contrast to standard Bayesian techniques, the artificial model can be robustified so that in misspecified models, decisions about the path of the (pseudo-true) parameter remain as good as in a corresponding correctly specified model.
AB - The paper investigates inference in non-linear and non-Gaussian models with moderately time-varying parameters. We show that for many decision problems, the sample information about the parameter path can be summarized by an artificial linear and Gaussian model, at least asymptotically. The approximation allows for computationally convenient path estimators and parameter stability tests. Also, in contrast to standard Bayesian techniques, the artificial model can be robustified so that in misspecified models, decisions about the path of the (pseudo-true) parameter remain as good as in a corresponding correctly specified model.
UR - http://www.scopus.com/inward/record.url?scp=77955739998&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77955739998&partnerID=8YFLogxK
U2 - 10.1111/j.1467-937X.2010.00603.x
DO - 10.1111/j.1467-937X.2010.00603.x
M3 - Article
AN - SCOPUS:77955739998
SN - 0034-6527
VL - 77
SP - 1508
EP - 1539
JO - Review of Economic Studies
JF - Review of Economic Studies
IS - 4
ER -