Friday, April 12, 2013

This Week's Reading

This past week I've been somewhat pre-occupied with the final exams for my undergraduate Economic Statistics course, and graduate Econometrics, courses. However, I've still managed to get some reading done, including the following miscellaneous papers:

  • S. Amini et al., 2012. Fixed versus random: The Hausman test four decades later. Unpublished paper, Dept. of Finance, Virginia Polytechnic & State University.
  • J. D. Angrist, 2001. Estimation of limited dependent variable models with dummy endogenous regressors. Simple strategies for empirical practice. Journal of Business and Economic Statistics, 19, 2-28.
  • A. Chesher & A. Rosen, 2013. What do instrumental variables deliver with discrete dependent variables? Cemmap Working Paper CWP 10/13, Dept. of Economics, University College London.
  • G. Chortareas & G. Kapetanios, 2013. How puzzling is the PPP puzzle? An alternative half-life measure  of convergence to PPP. Journal of Applied Econometrics, 28, 435-457.
  • S. N. Durlauf et al., 2012. Is God in the details? A reexamination of the role of religion in economic growth. Journal of Applied Econometrics, 27, 1059-1075.
  • C. Feng et al., 2013, Geometric mean of nonnegative random variable. Communications in Statistics - Theory & Methods, in press.
  • R. N. Rodriguez, 2013. Building the big tent for statistics. Journal of the American Statistical Association, 108, 1-6.
  • T. Wozniak, 2012. Testing causality between two vectors in multivariate GARCH models. Research Paper 1139, Dept. of Economics, University of Melbourne.


© 2013, David E. Giles


  1. Dear Professor Dave,

    I asked you before about the runing of VAR model, I have a question now about the impulse response functions. After runing a var and finding some coefficent not statistically significant, could I interpret their IRF?
    Could I replace the IRF by the OIRF to have better result?

    Thank you very much

  2. I'd be very cautious of using any IRF if you have concerns about the underlying VAR - e.g., lack of significance, serial correlation, etc.

  3. The first paper on the Hausman test seems odd to me: why do some econometricians say that the Hausman test tells us whether to use random or fixed effects, while some statisticians say that it does no such thing (because there is a very simple way to correct for endogeneity in random effects)? See, for example. Also see Bell/Jones paper, "Explaining Fixed Effects: Random Effects modelling of Time-Series Cross-Sectional and Panel Data."