Journal of Monetary Economics 41 (1998) 533
Download 72.51 Kb. Pdf ko'rish
|
debucyguide
F. Canova / Journal of Monetary Economics 41 (1998) 533– 540
models that Burnside uses are a misspecified description of the data? The exercise does not give us any hint of what is the power of the proposed test in this situation. Moreover, the exercise sheds no light on what would happen when data from different misspecified models is filtered with alternative filters because it is not hard to think of situations where some models may be more mispecified than others at certain frequencies, either because the transmission mechanism of shocks is too rudimental or because certain models are build with an eye at specific (stylized) facts previously spotted in the data. This may be an area of interesting research for the future, but the tools used by Burnside would not help us make progress in this direction. Wald tests are inappropriate if there is no available model which may have generated the data (roughly, there is an unknown non-centrality parameter which biases testing towards rejection). My guess is that Wald tests are even more inappropriate when filtered data from misspecified models are used in constructing the tests. Note that this problem is entirely independent of how parameters are estimated (the additional problems which may arise with GMM estimation are well known to Burnside and all of those who have contributed to increase our understanding of the mechanics of this estimation approach, see, e.g. special issue of the Journal of Business and Economic Statistics of July 1996 devoted to this question). To conclude, the Monte Carlo exercise conducted by Burnside has very little bearing with the two main questions that applied business cycle researchers are concerned with, document features of actual business cycle and evaluate the fit of misspecified models to the filtered data. Having said so, I have to admit that I do not have a fully articulated and comprehensive alternative able to address both issues. The silence of the paper, especially on the second problem, is partly a reflection of this ignorance. I believe that the idea of analyzing the effect of filtering the data and of systematically recording the resulting differences, together with some of the testing ideas I have suggested in Canova (1994) and (1995), constitute the skeleton for an interactive empirical approach which is more sensible than the one suggested by Burnside. Acknowledgements I would like to thank Morten Ravn, Russell Cooper and Adrian Pagan for endless conversations on the issues discussed in the paper. References Beveridge, S., Nelson, C., 1981. A new approach to decomposition of economic time series into permanent and transitory components with particular attention to measurement of the business cycle. Journal of Monetary Economics 7, 151—174. F. Canova / Journal of Monetary Economics 41 (1998) 533– 540 539 Blanchard, O., Fisher, S., 1989. Lectures on Macroeconomics. MIT Press, Cambridge, MA. Canova, F., 1994. Does detrending matter for the determination of the reference cycle and for the selection of turning points? Economic Journal, forthcoming. Canova, F., 1995. Sensitivity analysis and model evaluation in dynamic general equilibrium economies. International Economic Review 36, 477—501. Evans, P., Reichlin, L., 1994. Information, forecast and the measurement of the business cycle. Journal of Monetary Economics 33, 233—254. Hamilton, J., 1989. A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica 57, 357—384. Journal of Business and Economic Statistics (1996) Special Issue on GMM Estimation, July. King, R., Watson, M., 1996. Money, prices, interest rates and the business cycle. Review of Economics and Statistics LXXVIII, 35—54. Kydland, F., Prescott, E., 1990. Business cycles: real facts and monetary myth. Federal Reserve Bank of Minneapolis, Quarterly Review, Spring 3—18. Ravn, M., Ulhig, H., 1997. On adjusting the HP filter for the frequency of the observations. Tilburg University Working Paper. Watson, M., 1986. Univariate detrending methods with stochastic trends. Journal of Monetary Economics 18, 49—75. 540 F. Canova / Journal of Monetary Economics 41 (1998) 533– 540 Download 72.51 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling