Comparing Predictive Accuracy, Twenty Years Later: A Personal Perspective on the Use and Abuse of Diebold-Mariano Tests
NBER Working Paper No. 18391
---- Acknowledgements ----
This paper is based on a lecture delivered in March 2012 at CREATES, University of Aarhus, Denmark. For valuable comments I am grateful to participants at the CREATES lecture, especially Niels Haldrup, Soren Johansen, Asger Lunde, and Timo Terasvirta. I am also grateful to participants at Penn's Econometrics Lunch Seminar, as well as Peter Hansen, Lutz Kilian, Barbara Rossi, Frank Schorfheide, Minchul Shin, Norman Swanson, Allan Timmermann, Ken Wolpin and Jonathan Wright. The usual disclaimer most definitely applies. The views expressed herein are those of the author and do not necessarily reflect the views of the National Bureau of Economic Research.