In this article, we use the unrestricted two-regime autoregressive threshold model to test both nonlinearity
and stationarity of China’s real exchange rate against its Hong Kong and Macau special administrative
regions (SARs). Our main finding is that China’s real exchange rate is neither linear nor stationary, indicating
that the purchasing power parity does not hold between China Mainland and its two SARs, which implies, to
certain extent, the three economies may not meet the condition of constituting an optimal currency area.
ZHENG Xiaoya1*(郑晓亚), YOU Haibo2 (尤海波)
. Do China Mainland and SARs Constitute an Optimal Currency Area? Evidence from Nonlinearity and Stationarity Behavior Testing on Real Exchange Rates[J]. Journal of Shanghai Jiaotong University(Science), 2016
, 21(3)
: 328
-334
.
DOI: 10.1007/s12204-016-1729-9
[1] MUNDELL R A. A theory of optimum currency areas[J]. American Economic Review, 1961, 51(4): 657-665.
[2] NARAYAN P K, NARAYAN S. Are real exchangerates nonlinear with a unit root? Evidence on PPPfor Italy: A note [J]. Applied Economics, 2007, 39(19):2483-2488.
[3] TAYLORM P, PEEL D A, SARNO L. Nonlinear meanreversion in real exchange rates: Towards a solution tothe purchasing power parity puzzles [J]. InternationalEconomic Review, 2001, 42(4): 1015-1042.
[4] CHORTAREAS G, KAPETANIOS G. The yen realexchange rate may be stationary after all: Evidencefrom nonlinear unit root tests [J]. Oxford Bulletin ofEconomics and Statistics, 2004, 66(1): 113-131.
[5] CANER M, HANSEN B E. Threshold autoregressionwith a unit root [J]. Econometrica, 2001, 69(6): 1555-1596.
[6] KILIAN L, TAYLOR M P. Why is it so difficult tobeat the random walk forecast of exchange rates? [J].Journal of International Economics, 2003, 60(1): 85-107.
[7] SEKIOUA S H. Nonlinear adjustment in the forwardpremium: Evidence from a threshold unit root test [J].International Review of Economics and Finance, 2006,15(2): 164-183.
