Professional Documents
Culture Documents
Two series x and y are not stationary but linear combination of these series is stationary. Then these
series x and y are co-integrated.
Z=Q1 X+Q2 Y
If in short tern there is short term disequilibrium than we will see that in how much time this
disequilibrium is corrected in how much time.
There can be theory in back of this co-integration. Or this theory may not be there.
If co-integration exists than Error Correction Method ECM will be used and in this Granger casualty test
will be used (uni directional granger causality- bi directional casualty)
if co-integration is not there that VAR model will be used. And VDA
Yt=β0+β1yt-1+β2 t-1+β3t
t-Statistic Prob.*
Selected
(0.05 level*)
Number of
Cointegrating
Relations by
Model
Information
Criteria by
Rank and
Model
Log
Likelihood by
Rank (rows)
and Model
(columns)
0 29907.68 29907.68 29909.94 29909.94 29910.93
1 29910.70 29910.70 29911.61 29911.93 29912.87
2 29911.60 29911.61 29911.61 29913.56 29913.56
Akaike
Information
Criteria by
Rank (rows)
and Model
(columns)
0 -12.41458 -12.41458 -12.41469* -12.41469* -12.41427
1 -12.41417 -12.41376 -12.41372 -12.41343 -12.41341
2 -12.41288 -12.41206 -12.41206 -12.41203 -12.41203
Schwarz
Criteria by
Rank (rows)
and Model
(columns)
0 -12.37150* -12.37150* -12.36892 -12.36892 -12.36580
1 -12.36571 -12.36395 -12.36256 -12.36093 -12.35956
2 -12.35904 -12.35552 -12.35552 -12.35280 -12.35280
All answer are zero so we will use third will be used now.
LAUX LAUD
0.246522 4.745207
2.307867 -3.959400
LAUX LAUD
C 0.000219 1.44E-05
(0.00186) (0.00105)
[ 0.11811] [ 0.01376]
D(LAUX) D(LAUD)
D(LAUX(-1)) -0.042838 -0.003632
(0.01451) (0.00820)
[-2.95204] [-0.44270]
C 0.000444 2.97E-05
(0.00021) (0.00012)
[ 2.12234] [ 0.25101]
Varian
ce
Decom
position
of
D(LAU
X):
Period S.E. D(LAUX) D(LAUD)
Varian
ce
Decom
position
of
D(LAU
D):
Period S.E. D(LAUX) D(LAUD)
Choles
ky
Orderin
g:
D(LAU
X)
D(LAU
D)