Vector Error Correction Estimates
Date: 03/11/12 Time: 20:12
Sample (adjusted): 1993 2010
Included observations: 18 after adjustments
Standard errors in ( ) & t-statistics in [ ]
Cointegrating Eq: CointEq1 CointEq2 CointEq3
LOG_IR(-1) 1.000000 0.000000 0.000000
LOG_IP(-1) 0.000000 1.000000 0.000000
LOG_PG(-1) 0.000000 0.000000 1.000000
LOG_SZ(-1) -3.458571 -2.858526 -1.362368
(0.28273) (0.22519) (0.10925)
[-12.2326] [-12.6939] [-12.4697]
RZ(-1) -0.633781 -0.522529 -0.279406
(0.08507) (0.06776) (0.03287)
[-7.44969] [-7.71154] [-8.49914]
C 23.02698 16.47938 5.285923
上面中为什么有这么多1.0000?
Roots of Characteristic Polynomial
Endogenous variables: LOG_IR LOG_IP LOG_PG LOG_SZ RZ Exogenous variables:
Lag specification: 1 1
Date: 03/11/12 Time: 20:19
Root Modulus
1.000000 1.000000
1.000000 1.000000
-0.487227 - 0.756865i 0.900131
-0.487227 + 0.756865i 0.900131
0.743908 - 0.387911i 0.838972
0.743908 + 0.387911i 0.838972
0.469927 - 0.593432i 0.756962
0.469927 + 0.593432i 0.756962
-0.393595 0.393595
0.226310 0.226310
VEC specification imposes 2 unit root(s).
那这个VEC模型还有效吗?