: Perron,Zivot Andrew (1992), Nelson Plosser (1982),, 5 %4 ;, 5 % 3,Lumsdaine Papell (1997) Zivot Andrew (1992), Zivot Andrew (1992),, Perron (1989),,

Size: px

Start display at page:

Download ": Perron,Zivot Andrew (1992), Nelson Plosser (1982),, 5 %4 ;, 5 % 3,Lumsdaine Papell (1997) Zivot Andrew (1992), Zivot Andrew (1992),, Perron (1989),,"

晶莨徐
5 years ago
Views:

1 :,10, 6, GDP GDP 1 2,, : ( ) (segmented2trend2stationary),,,,dicky Fuller (1979),(persistence), Nelson Plosser (1982) 14, 13,,Perron (1989),,Nelson Plosser (1982) 1929,Perron 14 11,Perron,1973,,,,Perron, (Christiano, 1992 ; Zivot Andrews, 1992) 3, edu. cn ;,,, edu. cn, ( ) (MEXT Grant2in2Aid for JSPS Fellows) 11

2 : Perron,Zivot Andrew (1992), Nelson Plosser (1982),, 5 %4 ;, 5 % 3,Lumsdaine Papell (1997) Zivot Andrew (1992), Zivot Andrew (1992),, Perron (1989),,,,, ( Perron, 1989, 1990 ; Perron Vogelsan, 1992), Li (2000, 2005) Li (2000) , ( GDP), Perron (1997),, (Smyth Inder, 2004),Li (2005), , Smyth Inder (2004) GDP,,,,, Li Smyth Inder, 10,,, ( Zivot Andrews, 1992 ; Lumsdaine Papell, 1997),,,,,, : ; ;,,; Ohara (1999),, 1,,Li (2005) (),Li, 12

3 , China s National Income, ,Belson Plosser (1982) GDPGDP, 1952 GDP GDP GDP, GDP GDP GDP (implicit GDP deflator), 1952 GDP ; GDP GDP ;(L), ; (W), (RW) 1952 (CPI) ; (BC) (DL),1952 CPI ( I), 1990, Hsueh Li (1999), (1952 = 100), (C) (T) 1952 CPI ( ),,, X i ADF k X i, t = i + i t + i, t X i, t c i, j X i, t - j + e i, t (1), X i, t i t ( t = 1,2,, T) ; e i, t (0, ) (1) X i, t - j, BLUE() k k k k GDP GDP, ,,,, Hsueh Li (1999), ( ) (1990 ) , (1) 13

4 :, t2sig (Perron, 1989 ; Hall, 1994) t2sig k max, k max, k,c i, k t 10 % 116 k max 6,c i, kmax t,k ( Perron, 1989 ; Zivot Andrews, 1992 ; Lumsdaine Papell, 1997) X i t i, ( X i, t ), 5000,(1) i i t^, t^ GDP GDP, (ARMA) (Zivot Andrews, 1992), BIC (Schwarz, 1978) p q ( p, q 5), X i, t = e i, t (Lumsdaine Papell, 1997) Zivot Andrews (1992),k,,, i t^ k 1 ADF PP KPSS ADF ( t2sig ) 1 % 5 % 10 % GDP GDP : % 5 % 10 %,ADF PP KPSS, 1 4 ADF ( BIC ), ; PP ; KPSS, t - sig ADF, t - sig ADF, t2sig,, AIC BIC (Ng and Perron, 1995 ; Perron, 1997) 14

5 ( ), ,,,, { X i, t }, { X i, t } TB Zivot Andrews (1992) A C A, C, X i, t = i + i t + i DU t + i, t X i, t - 1 k + c i, j X i, t - j + e i, t (2) k X i, t = i + i t + i DU t + i DT t + i, t X i, t c i, j X i, t - j + e i, t (3), DU t DT t TB,DU t = 1 ( t > TB), DT t = ( t - TB ) 1 ( t > TB ), (0115T T) (2) (3), X i, t - 1 i, t t TB,, (0115T T) (2) i (3), t^, 5000 i t^ 2 A C A, C 5 %, 4, C A,C, 1989, ( steady2state) ( ), %, %, %,, 1989, 1147 %, %,1978,, 1 %,, 1982,,, 1960,, 1 %,,,,, (8) (8), 15

6 : 2 A TB t 1 % 5 % 10 % t k GDP GDP C TB t 1 % 5 % 10 % t t k GDP GDP : % 5 % 10 % ( ) Lee Strazicich (2003),,, (Lumsdaine Papell, 1997 ; Arestis Mariscal, 1999 ; Ben2David Papell, 2000),Arestis Mariscal (1999) 26 OECD,25,19 Lumsdaine Papell (1997) Zivot Andrews (1992),AA CC : 16

7 X i, t i, t = i + i t + i DU 1, t = i + i t + i DU 1, t + i DU 2, t + i DT 1, t + i, t X i, t i DU 2, t k + c i, j X i, t - j + e i, t (4) k + i DT 2, t + i, t X i, t c i, j X i, t - j + e i, t (5) DU 1, t DU 2, t TB1 TB2 ( TB2 > TB1 + 1), DT 1, t DT 2, t,du 1, t = 1 ( t > TB1), DU 2, t = 1 ( t > TB2), DT 1, t = ( t - TB1) 1 ( t > TB1), DT 2, t = ( t - TB2) 1 ( t > TB2) (TB2 > TB1 + 1), X i, t - 1 i, t t,tb1 TB2, 3 AA CC AA GDPGDP, 5 %, CC GDPGDP, 1 %,,,Lumsdaine Papell (1997, p1 217) : ( ),, CC ( ) 5 %,CC, CC,GDP GDP 1960,, 1 %, ,,,,GDP GDP ,, GDP GDP 8151 % 7129 %, % ( = (8151 % %)Π6194 %) % ( = (7129 % %)Π4174 %) C,CC 1963, C, 1989 ; C, 1960, 1994 ; C, ; C, ,,, CC (Dolmas,1999 ; Lumsdaine Papell, 1997),, 3 C CC,, 17

8 : 3 AA TB1 TB2 t 1 % 5 % 10 % t g t g k GDP GDP CC TB1 TB2 t 1 % 5 % 10 % t g t g t t k GDP GDP : % 5 % 10 % Li (2000, 2005) Li (2000),,,,,,A AA,A, AA,,,,,,,,, (6) C CC,A AA, 18

9 m1 DY t = a 0 + a 1 i DY t - i i = 1 m2 + a 2 i DI t - i i = 1 + u t (6 - A) n1 n2 DI t = b 0 + b 1 j DI t - j + b 2 j DY t - j + e t (6 - B) DY t DI t ; u e ; t ; m n Hendry General2to2Specific a 2 i = 0, ; b 2 j = 0,,, Perron (1994) EX i, t EX i, t = X i, t - TR i, t (7) TR i, t X i, t OLS p X i, t = a i + b i t + g i,1 DU 1, t + g i,2 DU 2, t + h i,1 DT 1, t + h i,2 DT 2, t + c i, j X i, t - j + i, t X i, t,(8) C, (3) ; X i, t, (8) CC,(5) (3) (5) (8) p k, (8) : TR i, t = p ^a - ^b j^c i, j p ^c i, j ^b p 1 - p 2 t + (1 - ^c i, j ^c i, j L i, j ) ( m = 1 ^g i, m DU m, t (8) 2 + ^h i, n DT n, t ) n = 1 (9) (7), ( ) (4),,,1 % ,, GDP (9) 19

10 : GDP 6,,,,,, ( ),,,,,,,,,,,,,,, ;, 4 General2to2Specific L GDP ; 4 I GDP ; 4 GDP L ; 1 GDP I ; 3 BC GDP ; 2 DL GDP ; 3 GDP BC ; 4 GDP DL ; 1 : 1,t ;1,F % 5 % 10 % Lumsdaine Papell (1997),,,, Lumsdaine Papell,50,3?,,,,,1999 :, Arestis, P., and Mariscal, I. B.,1999, Unit Root and Structural Breaks in OECD Unemployment, Economics Letter, 65, Ben2David, D., Papell, D.,2000, Some Evidence of the Continuity of the Growth Process among the G7 Countries, Economic Inquiry, 38, Christiano, L1,1992, Searching for a Break in GNP, Journal of Business and Economic Statistics, 10, Dickey, D1, Fuller, W1,1979, Distribution of the Estimators for Autoregressive Time Series with a Unit Root, Journal of the American Statistical Association, 74,

11 Dolmas, J1, Beldav, R., and Slottje, D.,1999, The U. S. Productivity Slowdown : a Peak through the Structural Break Window, Economic Inquiry, 37, Hall, A. D.,1994, Testing for a Unit Root in Time Series with Pretest Data Based Model Selection, Journal of Business and Economic Statistics, 12, Hsueh, T., and Li, Q.,1999, China s National Income, Westview Press. Lee, J., and Strazicich, M.,2003, Minimum Lagrange Multiplier Unit Root Test with Two Structural Breaks, Review of Economics and Statistics, 85, Li, X. M., 2000, The Great Leap Forward, Economic Reforms, and the Unit Root Hypothesis : Testing for Breaking Trend Functions in China s GDP Data, Journal of Comparative Economics, 28, Li, X. M., 2005, China s Economic Growth : What do We Learn from Multiple2break Unit Root Tests? Scottish Journal of Political Economy, 52, Lumsdaine, R., and Papell, D.,1999, Multiple Trend Breaks and the Unit2root Hypothesis, The Review of Economics and Statistics, 79, Nelson, C. R., and Plosser, C., 1982, Trends and Random Walks in Macroeconomic Time Series : Some Evidence and Implications, Journal of Monetary Economics, 10, Ng, S., and Perron, P.,1995, Unit Root Test in ARMA Models with Data2dependent Methods for the Selection of the Truncation Lag, Journal of the American Statistical Association, 90, Ohara, H., 1999, A Unit Root Test with Multiple Trend Breaks : a Theory and an Application to US and Japanese Macroeconomic Time Series, Japanese Economic Review, 50, Perron, P. 1989, The Great Crash, the Oil Price Shock, and the Unit Root Hypothesis, Econometrica, 57, Perron, P.,1990, Testing for a Unit Root in a Time Series with a Changing Mean, Journal of Business and Economic Statistics, 8, 153 Perron, P.,1994, Trend, Unit Root and Structural Change in Macroeconomic Time Series. In B. Bhaskara Rao, Ed., Co2integration for the Applied Economists, pp New York : St. Martin s. Perron, P.,1997, Further Evidence on Breaking Trend Functions in Macroeconomic Variables, Journal of Econometrics, 80, Perron, P., and Vogelsan, T.,1992, Testing for a Unit Root in a Time Series with a Changing Mean : Corrections and Extensions, Journal of Business and Economic Statistics, 10, Smyth, R., and Inder B.,2004, Is Chinese Provincial Real GDP per capita Non2stationary? Evidence from Multiple Trend Break Unit Root Tests, China Economic Review, 15, Zivot, E., and Andrews, D.,1992, Further Evidence on the Great Crash, the Oil2price Shock, and the Unit2root Hypothesis, Journal of Business and Economic Statistics, 10, Structural Change of Chinese Macroeconomic and Financial Series and Causality Analysis Liang Qi and Teng JianZhou (Nankai University ; Northeast Normal University, Hitotsubashi University) Abstract :This paper applies unit root tests to 10 Chinese macroeconomic and financial series that allow for the possibility of up to two endogenous structural breaks. We find that 6 out of 10 series, which are GDP, GDP per capita, total number of employed persons, bank credit, deposit liabilities and fixed capital, can be more accurately characterized as a segmented trend stationary process around one or two structural breaks as opposed to a stochastic unit root process. The conclusions have important implications for policy2makers to formulate long2term economic growth strategy and short2run stabilization policies, as well as causality analysis among series. Key Words :Macroeconomic Time Series ; Unit2root Testing ; Multiple Breakpoints ; Segmented Trend Stationary ; Causality JEL Classification :C220 ;O110 ; E300 ( : ) (: ) 21

12 : A A 22

1970 Roulac (1996) (shock) (structure change) Barras and Ferguson (1985) Barras (1994) (1990) (1996) (1997) 1

1970 Roulac (1996) (shock) (structure change) Barras and Ferguson (1985) Barras (1994) (1990) (1996) (1997) 1 (1998) 1990 (Unit Root Test) (Cointegration) (Error Correction Model) 1 (1996) 2 (1990) 2 Barras