Revisiting the Out-of-sample Exchange Rate Predictability in the Monetary Model
2011/05/31
研討會日期 : 2011-05-31
時間 : 14:30
主講人 : Profeossor Hsiu-Hsin Ko
地點 : Conference Room B110
演講者簡介 : Hsiu-Hsin Ko got her Ph.D. in Economics from The Ohio State University in 2009. She is currently serving as Assistant Professor of Department of Applied Economics at National University of Kaohsiung. Her research field is in international macroeconomics, monetary economics, and applied econometrics.
演講摘要 : We utilize Monte Carlo simulations to evaluate, in finite samples, the forecasting performance of the monetary model. In each of experiment, the data generating process (DGP) is based on the assumptions of Engel and West (2005) for the present-value model. The simulated data are applied in the long-run regression test to evaluate the forecasting performance of the monetary model against the random walk model. While simulation results exhibit some evidence of exchange rate predictability, the evidence is not strong. The forecasting power of the monetary model depends on whether the random walk null hypothesis includes a drift. The simulation results based on the asymptotic distribution and the bootstrap percentile distribution differ in the aspect of exchange rate predictability across forecast horizons. The results from the asymptotic distribution show that the monetary model tends to outperform the random walk model at longer forecast horizons rather than at the short-run horizon. However, the tests based on the bootstrap distribution show the opposite results.
