Вы находитесь на странице: 1из 63

-

-
( )

2003

I.
EViews
......................................................................................................... 3
1. EViews ............................................................ 3
1.1. EViews ........................................................ 4
......................................................... 9
1.2. EViews ........................................... 9
....................................................... 12
2. ..................................................... 12
2.1. ....................................................... 12
....................................................... 16
2.2 ................................................ 17
....................................................... 20
2.3. ...................................... 20
....................................................... 23
2.4. ...................... 23
....................................................... 31
2.5. (
)..................... 31
....................................................... 33
2.6. ARIMA( p, d , q ) .............................. 33
....................................................... 45
2.7. ( -) .......................... 45
....................................................... 47
2.8. ARMA(p, q)....................... 48
....................................................... 51
3. ................................................... 51
3.1. (
) ....................................................................................................................... 51
....................................................... 51
3.2. ...................... 52
....................................................... 54
3.3. ..................................................................... 54
3.4. ................................................... 57
3.5. ...................................................... 58
3.3-3.5 ........................ 62



EViews 4.1 .
,
:

.
,
EViews (1), (2)
(3).
, EViews
.
,
,
ARIMA(p, d, q),
.
, ,
,
, ,
.
1. EViews
EViews :
EViews

File Edit Object View Procs Quick Options Window Help

Welcome to EViews

Path=c:\eviews\

DB=macro

WF=none

. Eviews Eviews
, .
Eviews 9
(File, Edit, Object, View, Procs,

Quick, Options, Window, Help),


,
.
. EViews ,
, , ,
.
Enter .
() EViews,
. , ,
, , ,
..

Eviews.
1.1. EViews
1. , ,
,
File/New/Workfile, :
Workfile Range
Frequency

Annual

Weekly

Semi-annual

Daily [5 day weeks]

Quarterly

Daily [7 day weeks]

Monthly

Undated or irregular

Range
Start date
94:1

OK

Cancel
End date
2003M6

:
, (,
, , , ,
()) .
(Annual). (Start date) (End date)
,
19**. , 1994 2002 .,
1994 94.

, 2002.
(Quarterly):
19** .
, 3- 1998 1998:3 98:3
1998Q3. , (4

), Q
( 1 4).
(Monthly). :
M ( 1 (
01) 12). , 1998 1998:08,
1998:8, 98:08 1998M8.
(Semi-annual):

S, , 94:1 ( 1994 ) 2003S1
( 2003 ).
(Weekly Daily).
,
()
:: (mm:dd:yyyy).
07:10:1974,
, 10 1974
, , 10 1974.
(Undated or irregular).
( ),
.
,
94 2002 ,
, :
1994-2002
1994:1-2002:2
1994:1-2002:4
1994:01-2002:12
01/01/1994-12/28/2002
01/01/1994-12/31/2002
94-2002

,
, , EViews
( ) :

Wokrfile: UNTITLED
View Procs Object
Save Label+/Range: 1994:01 2003:06
Sample: 1994:01 2003:06

Show Fetch Store Delete Genr Sample


Filter: *
Default Eq: None

c
resid

, Range ,
, Sample
. 1.2
EViews.
2. EViews
. ,
, Microsoft Excel .
EViews File/Import/Read Text-LotusExcel
? X

Open
Look in:

C:\

File name
Files
type

Open
of

Excel(*.xls)

Cancel

Update default directory

Microsoft Excel, ,
. ,
Excel Spreadsheet Import

Data order

Upper-left data cell

B2

By Observation series in columns

E
E
m
Exxxccceeell5
l55+++ssshhheeeeeettt nnnaaam
meee

By Series series in rows


Names for series or Number if named in file
5

6
Import sample
1994:01 2003:06

Reset sample to:


Current sample
Workfile range

E
E
Exxxpppooorrrttt oooppptttiio
ioonnnsss
W
r
i
t
e
d
a
t
W
Wrriittee ddaatteee///ooobbbsss

E
V
w
m
E
EV
Viie
ieew
wsss dddaaattteeefffooorrrm
maaattt

FFFiirirrsssttt cccaaalle
leennndddaaarrr dddaaayyy

LLLaaasssttt cccaaalle
leennndddaaarrr dddaaayyy
W
W
m
Wrrriititteeessseeerrriie
ieesss nnnaaam
meee
A
S
C
I
I
T
e
x
t
d
e
l
i
m
i
t
A
AS
SC
CIIII--TTeexxtt ddeelliim
miitteeerrr

TTTaaabbb

S
S
Spppaaaccceee

C
m
m
C
Cooom
mm
maaa

To end pf range
OK

Cancel

,
Microsoft Excel,
EViews, , .
,
,
EViews. -,
Microsoft
Excel 4.0, , Microsoft Excel.
Excel5+sheet name.
Microsoft Excel, ,
, , .
, ,
.
-, ,
, Microsoft Excel (

), , ,
. ,
Microsoft Excel ,
EViews
, Microsoft Excel,
EViews.

-, EViews
Microsoft Excel. , , Microsoft
Excel, ,
.
(Data order). Microsoft Excel
, By Observation
series in columns ( ). ,
, By Series series in rows.
,
(Upper-left data cell).
, .
2, ,
, ,
.
Microsoft Excel (Excel5+sheet name).
Microsoft Excel,
,
Microsoft Excel. Microsoft Excel
4.0, .
()
, Microsoft Excel
(Names for series or Number if named in file).
Microsoft Excel,
()
, :
, .
Microsoft Excel,
,
.
(Import sample).
, ,
() , (
, ,
, EViews NA).
,
OK, EViews,
, , :

Wokrfile: UNTITLED
View
Procs
Object
Range: 1994:01 2003:06
Sample: 1994:01 2003:06

Save

Label+/Show
Filter: *

Fetch

Store
Delete
Genr
Default Eq: None

_
Sample

resid

y1

y2

y3

y1, y2, y3 .

1. EViews 3.1 EViews 4.1


2. EViews,

1000

1.2. EViews
.
EViews Windows,
, , Windows,
, Microsoft Word, Excel ..
EViews .
EViews.
, EViews,
File/Save File/Save as ,
Save (
).
, ,
File/Open EViews
.

.
EViews 1,
() :
@all , ;
@first ;
@last .
.
, .
,
smpl -1 -2
,
1993 1998 .

smpl @all
smpl @first 1996:12
smpl 1995:08 @last,
, ,
,

1996 ., , ,
1995 . .
-
,
, , :
smpl 1993:09 1997:1.

1993 1997 .
, EViews
.
, 1993 1997 .

smpl @first+9 @last-10.
.
Sample
,

, , EViews, @.

10

Sample

Sample range pairs (or sample object to copy)


2
1993:01 1998:07
OK

IF condition (optional)

Cancel

,
.
Quick EViews.
Quick
EViews Sample,
.
EViews
EViews
(),
:
(Coefficient Vector)


(Equation)
=

(Graph)
(Group)
G

(LogL)
#
() (Scalar)
() (Series)

(State space)
ss
S
(System)
(SYM Symmetric Matrix)

(Matrix)
[]
(Model)

(Pool)
P
(Sample)
Table
(Table)
TXT
(Text)
VAR
2

,
.

11

(VAR Vector Autoregression)


( ) (Vector/Row Vector)

,
Object/New Object EViews

New Object
Type of object

Name for object

Sample

Untitled

Equation
Graph
Group
LogL
Matrix-Vector-Coef
Model

OK

Pool
Sample
Series
Sspace

Cancel

System
Table
Text
VAR

,
OK, , ,
,
(, ),
, .

1.
2.

, ,
DEMO.wf.

y1, y2, y3, y4, y5.

2.
2.1.
EViews
, EViews.
, ,
.

12

3 :
Series: y1

Workfile: CP-2

View Procs Object

Print Name Freeze

Transform Edit+/- Smpl+/- Label+/- Wlde+- InsDel Title Sample Genr


Y1
Last updated: 03/20/03 - 12:31

Modified: 1 1000 // y1=-2


Modified: 2 1000 // y1=0.5*y1(-1)+nrnd
1

-2

-0.9566483

1.3842802

1.0306127

-0.3917438

-1.6905064

View :
SpreadSheeet ;
Graph (
. 2.2);
Descriptive Statistics
: , .. (
. 2.3);
Tests for Descriptive Statistics
( . 2.4);
Distribution ,
;
One-Way Tabulation
, :
,
,
, %, ,
( %);
Correlogram (
. 2.5);
Unit Root Test
( . 2.7);
BDS Independence Test
.
,

, -
( ),
.

13

. :
, ;
Conversion Options EViews

;
Label , :
, ,
, , , , .
Procs :
(Generate by Equation
Genr );
(Resample),
;
(Seasonal Adjustment )
: Sensus X12; X11
(Historical); Tramo/Seats; Moving Average Methods;

(Exponential Smoothing);
-
(Hodrick-Prescott Filter).
Object ,

:
Store to DB EViews;
Update from DB ,
EViews;
Copy Object ;
Name ( Name
);
Delete EViews (
Delete );
Freeze Output (
Freeze );
Print ( Print
);
View Options
, :
Sample +/-;
Edit +/-;
Label +/-;
Wide +/-;
InsDel;
Title
Print . Print
.
Name . Name
.
Freeze .
.

14

Transform4 .
(Level) :
1 Period Difference :
yt yt 1
1 Year Difference

:
yt yt 12

. ..
.
1 Period % Change :
yt yt 1
100%
yt


:
1 Period % Change (Annual Rate)

12

y y

t 1
100%
t

1
1
+

yt

1 Year % Change

:
yt yt 12
100% ,
yt

. ..
.
Edit+/- . /

. ,
Enter.
Smpl+/- .
,
(),
.
Label+/- . Label+/-
( )
, ..
Wlde+- . Wlde+-
( ).
InsDel .
/ .
, Insert
obs and move subsequent obs down one. ,
, ( ),
, .

EViews 3.1.

15

,
.
- , Delete obs
and move subsequent obs up one.
.
Title .
(
) :
.
Sample .
( . .1.2).
Genr .

,
EViews. ,

:
Generate Series by Equation

Enter equation
ind=

Sample
1994:01 2003:06

OK

Cancel

Sample ,
.

1. DEMO.wf :
wn=nrnd*
y1=a0+a1*y1(-1)+wn
y2=a0+a2*y2(-1)+wn,
a0, a1, a2 ,
,
.
* nrnd EViews ,
,

16

y1(-1) ,
.,
y1 y2, -,
( ,
, , Edit ).
-,
Generate Series by Equation
2 1000.
2. .
2.2
, ,
. View
Graph. :
5 (Line)
. ( ),
.

110
100
90
80
70
60
50
93

94

95

96

97

98

99

00

01

02

IN D

(Bar) .
,
.



(IND) 1993 . 2003 ( 1993 . =
100)

17

110
100
90
80
70
60
50
93

94

95

96

97

98

99

00

01

02

IN D

(Spike)
, (
).
110
100
90
80
70
60
50
93

94

95

96

97

98

99

00

01

02

IN D

(Seasonal Stacked Line)




. :
, , ..
. ,
.

18

IND by Season
110
100
90
80
70
60
50

Jan Feb Mar Apr May Jun


IND

Jul

Aug Sep Oct

Nov Dec

Means by Season

(Seasonal Split Line)




. .
.

IN D b y S eason
110
100

Jan
F eb
M ar
Apr
M ay
Ju n
Ju l
Aug
S ep
O ct
N ov
D ec

90
80
70
60
50
93

94

95

96

97

98

99

00

01

02

03

,
Freeze
. Name
, ( )
EViews.

Object/New Object/Graph EViews.
,
. ,
, () EViews.

19

graph,
.
()
graph < >.< ()> <
( )>

Enter.
EViews :
bar

errbar

,
.
,
. , ,
,
.

hilo

4- .
errbar.
.

line

( ,
)

pie

(
)

scat

xy, ( )

spike

xy

(x, y). (OX)


,
. ,
y(x)

xyline

xy

xypair

xy

1.

wn, y1,
y2, 2.2.

2.3.
Descriptive Statistics View
,
.
Histogram and Stats
Stats Table
Stats by Classification

20

(Histogram and Stats)


( ,
, )
24

Series: IND
Sample 1993:01 2002:02
Observations 110

20
16
12
8
4
0
50

60

70

80

90

Mean
Median
Maximum
Minimum
Std. Dev.
Skewness
Kurtosis

64.91535
62.88370
107.1960
50.82430
10.64339
1.744585
6.157356

Jarque-Bera
Probability

101.4897
0.000000

100

: (Mean),
(Median), (Maximum),
(Minimum), (Std. Dev.),
(Skewness) (Kurtosis), - (JarkeBera) - (Probability).

yt , .
( ) ,
,
/
, .
:
:
T

(y
t =1

T 1
;
:
3

1 y
S = t
,
T

T 1

T
,
. ,
, , .

21

,
, ;
:
4

1 T y
K = t
.
T t =1

3.

3,

3.
, , - ,
. - ,
,
:
T k 2 ( K 3)
S +
6
4

S , K ,
k ( ,
.. ). -
2 (2) .
2

JB =

(Stats Table).
( )
: (Mean),
(Median), (Maximum),
(Minimum), (Std. Dev.),
(Skewness) (Kurtosis), - (Jarke-Bera)
- (Probability), (Sum)
(Sum Sq. Dev.).

(Stats by Classification).
.
, ,
( , , ,
)
1995 .
1998 . 1998 . 2003 .

0, t 1998 : 08,
DTt =
1, .
View/Descriptive Statistics/Stats by Classification
:
Stats by Classification
Statistics
3

X
Series/Group for Classify

Mean

dt

Sum

NA handling

Median

Maximum

Output Layout
Table Display

Treat NA as category

3 Show Row Margins


3 Show Column Margins
3 Show Table Margins

22

Minimum

Group into bins if

Std. Dev.

# of values >

100

List Display

Avg. count <

S
S
w
S
M
Shhhooow
wS
Suuubbb---M
Maaarrrgggiin
innsss
S
p
a
r
s
e
L
a
b
e
l
s
S
Sppaarrssee LLaabbeellss

Quintile

000...555

Skewness
Kurtosis

Max # of bins

# of Nas
3

Observations
OK

Cancel

Options

,
:
Descriptive Statistics for IND
Categorized by values of DT
Date: 08/08/03 Time: 13:04
Sample: 1995:01 2003:05
Included observations: 101
DT
0
1
All

Mean
58.59149
65.57914
62.53502

Median
58.39150
66.40950
62.33490

Max
69.26350
75.67090
75.67090

Min.
51.70410
50.83030
50.83030

Std. Dev.
4.019442
5.516713
6.009558

Obs.
44
57
101

,
, Series/Group for Classify
, .

1. wn, y1, y2.


,
?
2.
.
?
2.4.
Eviews
. View/Tests for
Descriptive Stats/Simple Hypothesis Tests,
,
,
:
Series Distribution Tests
Test Value

X
Mean test assumption

Mean:

65

Median:

63

Mean test will use a known


standard deviation if supplied

Variance:

100

Enter s.d. if known

23

OK

Cancel

, ,
(
1993 .) 1993 . 2003 . 65, 63,
81. :
.
, m=65,
:

H0 : = m
HA : m
yt ,
t-:
m
,
t=

T
&& yt ,
. yt , t-
T-1 .
, EViews
z-:
m
z=

&& yt ,
, . yt
, z-
.

: ,
, 0,05, , 5%-
m ;
,
,
.
. ,
m=63, :
H 0 : med ( yt ) = m

H A : med ( yt ) m

EViews ,
.

24

Binomial sign test.

,
,
,
. EViews P-
, (
).
Wilcoxon signed ranks test. ,

,
( ). ,
, , , ,
. EViews -
t- (
ties) (correcting for both continuity and ties).
Van der Waerden (normal scores) test. ,
, . smoothed ranks.
, ,
.
. ,
=81,

H 0 : var( yt ) = 2

H A : var( yt ) 2

EViews :

2 =

(T 1) 2
2

, . yt
, 2 (T 1) .
,
,
.

:
Hypothesis Testing for IND
Date: 08/08/03 Time: 17:18
Sample: 1993:01 2003:05
Included observations: 125
Test of Hypothesis: Mean = 65.00000
Sample Mean = 65.66241
Sample Std. Dev. = 10.21339
Method
t-statistic

Value
0.725120

Probability
0.4697

Value
159.6894

Probability
0.0170

Test of Hypothesis: Variance = 81.000000


Sample Variance = 104.3133
Method
Variance Ratio

25

Test of Hypothesis: Median = 63.00000


Sample Median = 64.22410
Value
68
0.894427
1.749391
2.172257

Method
Sign (exact binomial)
Sign (normal approximation)
Wilcoxon signed rank
van der Waerden (normal scores)

Probability
0.3712
0.3711
0.0802
0.0298

Median Test Summary


Category

Count

Mean Rank

68
57
0

68.3529412
56.6140351

Obs > 63.00000


Obs < 63.00000
Obs = 63.00000
Total

125

, 5%-
,
1993 . 2003 . 65 ;
81 ;
.
, View/Tests for Descriptive Stats/Equality Tests by
Classification,
,
.
,
,
: 1995 . 1998
1998 . 2003 . ,
() () () (Series/Group for
classify),
: (Mean),
(Median) (Variance)
Tests By Classification
Series/Group for classify

X
NA handling

dt

Treat NAs as category

Test equality of

Group into bins of

Mean

# of values >

100

Median

Avg. count <

Max # of bins

Variance
OK

Cancel

. ,
(ANOVA)
: , ,

, ,

26


.
i- g as y g , i , i=1,, n g
g=1, 2,,G. (Between) (Within)
:
2

SS B = n g ( g )
G

g =1

ng

SSW = ( y g , i g ) ,
G

g =1 i =1

,
, g ,
g=1,, G. F-
:
SS B (G 1)
F=
,
SSW (T G )
. F- F-
(G-1, -G) , ,
o.
Test for Equality of Means of IND
Categorized by values of DT
Date: 08/11/03 Time: 13:33
Sample: 1995:01 2003:05
Included observations: 101
Method

df

Value

Probability

99
(1, 99)

7.073519
50.03467

0.0000
0.0000

Source of Variation

df

Sum of Sq.

Mean Sq.

Between
Within

1
99

1212.464
2399.015

1212.464
24.23248

100

3611.479

36.11479

Mean
58.59149
65.57914
62.53502

Std. Dev.
4.019442
5.516713
6.009558

Std. Err.
of Mean
0.605954
0.730707
0.597973

t-test
Anova F-statistic
Analysis of Variance

Total
Category Statistics
DT
0
1
All

Count
44
57
101

G=2, EViews t-,


F- T-G .
:

(Between) (Within)

27

Between =
Within =

SS B
df B

SSW
,
dfW

df B df W . F-

Between
F=
.
Within
. EViews ,
,
, ,
, .
,
.
, ,
, ,
.
Wilcoxon/Mann-Whitney signed ranks test. ,
,

- (. 2.3).
Chi-square test for the median. ANOVA
,
, .
(the median test). 2
2 (G 1) . EViews
(Yate), ,
.
Kruskal-Wallis one-way ANOVA by ranks.
- (Mann-Whitney test)
. ,
( 1,
). 1 2.
, .
Test for Equality of Medians of IND
Categorized by values of DT
Date: 08/12/03 Time: 13:02
Sample: 1995:01 2003:05
Included observations: 101
Method
Wilcoxon/Mann-Whitney
Wilcoxon/Mann-Whitney (tie-adj.)
Med. Chi-square
Adj. Med. Chi-square
Kruskal-Wallis

df

Value

Probability

1
1
1

5.948351
5.948351
30.60075
28.42072
35.42363

0.0000
0.0000
0.0000
0.0000
0.0000

28

Kruskal-Wallis (tie-adj.)
van der Waerden

1
1

35.42363
31.91995

0.0000
0.0000

Median
58.39150
66.40950
62.33490

> Overall
Median
8
42
50

Mean Rank
31.25000
66.24561
51.00000

Category Statistics
DT
0
1
All

Count
44
57
101

Mean Score
-0.617909
0.476983
-9.51E-17

EViews U-
( tie) -
. ,
. EViews 2 ( tie ).
2 (G 1) .
Van der Waerden (normal scores) test.
- ,
. EViews ,
2 (G 1) .

,
: ,
, (Count);
(Median);
, (> Overall Median);
6 (Mean Rank);

(Mean Score).
.

,
.
F-test (G=2).
s L2 s S2
. F-
:

F=

s L2
.
s S2

F- F- (TL 1, TS 1)
, .
6

:
(
1, ), , ,
, .

29

Siegel-Tukey test. ,
. ,
.
, (. ), .
,
1. 2, ,
3. , ,
4, 5, . EViews
-
.
Bartlett test. weighted average
variance . :
. ,
, 2 (G 1) .
, ,
. EViews
.
Levene test (ANOVA)
. F-statistic F (G 1, T G ) .

,

.
(robustness) .
Brown-Forsythe (modified Levene) test

Test for Equality of Variances of IND


Categorized by values of DT
Date: 08/12/03 Time: 18:07
Sample: 1995:01 2003:05
Included observations: 101
Method
F-test
Siegel-Tukey
Bartlett
Levene
Brown-Forsythe

df

Value

Probability

(43, 56)
1
(1, 99)
(1, 99)

1.883776
0.715720
4.623970
4.142851
3.613701

0.0263
0.4742
0.0315
0.0445
0.0602

Std. Dev.
4.019442
5.516713
6.009558

Mean Abs.
Mean Diff.
3.291818
4.448680
3.944700

Category Statistics
DT
0
1
All

Count
44
57
101

Mean Abs. Mean TukeyMedian Diff. Siegel Rank


3.291564
53.38636
4.411904
49.15789
3.923835
51.00000

Bartlett weighted standard deviation: 4.922649

30

7 (Category Statistics),
.

1. wn, y1, y2 ,

.
2.5. (
)

Correlogram View View/Correlogram:
X

Correlogram Specification
Correlogram of
Level
1st difference

OK

2d difference
Lags to include

Cancel

36

, (Correlogram
of) (Level), (1st difference)
(2d difference), 8,
9 (Lags to include).
, .
Series: Y1 Workfile: CP-2
View Procs Object Print Name Freeze

Date: 07/31/03 Time: 15:38


Sample: 1 1000
Included observations: 1000
Autocorrelation
.|****
.|**
.|*
.|*
.|
.|
.|
.|
.|

|
|
|
|
|
|
|
|
|

Sample Genr Sheet Stats Ident LineBar


Correlogram of Y1
5

Partial Correlation
.|****
.|
|
.|
|
.|
|
.|
|
.|
|
.|
|
.|
|
.|
|

1
2
3
4
5
6
7
8
9

AC

PAC

Q-Stat

Prob

0.548
0.313
0.162
0.094
0.045
0.027
-0.002
0.012
-0.003

0.548
0.017
-0.023
0.013
-0.012
0.006
-0.025
0.031
-0.022

301.56
399.80
426.25
435.08
437.14
437.85
437.86
438.00
438.01

0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000
0.000

.
, /4 ,
( ).
9
, ,
36 .
8

31

.|
.|
.|
.|

|
|
|
|

.|
.|
.|
.|

|
|
|
|

10
11
12
13

-0.012
-0.018
-0.022
0.007

-0.009
-0.006
-0.011
0.038

438.15
438.47
438.98
439.02

0.000
0.000
0.000
0.000

,
(Y1),
(CP-2). :
View ProcsObject

Print Name Freeze

SampleGenr Sheet Stats Ident Line Bar

Correlogram of Y1 ,
Y1.
(Date:
Time: ), (Sample: )
(Included observations: ).
Autocorrelation Partial Correlation

( ).
, k-
yt
T

rk =

(y

t = k +1

)( yt k )

(y
t =1

yt , yt .
k-

r1 ,

k 1
rk k 1, j rk j
kk = k =
j =1
,
k 1

1 k 1, j rk j

j =1

k =1
k >1

rk k-

k , j = k 1, j k k 1, k j ,
.


2
. k-
T
( )
, ,
() 5%-
.

32

AC PAC

, .
, , Q-Stat Prob Q-
- (Ljung-Box) P- . k
r j2

QLB (k ) = T (T + 2 )
j =1

Tj


k:
k

H0 :
HA :

r
j =1
k

2
j

r
j =1

2
j

=0
>0

-
yt , QLB (k ) 2 (k ) .
ARIMA( p, d , q ) , QLB (k )
2 (k p q ) .

1.

2.

DEMO.wf
(y3),
(y4=a0+wn+b1*wn(-1))
(1, 1) (y5).
wn, y1-y5, .
?
, , , ,
?

2.6. ARIMA( p, d , q )
EViews

ARIMA( p, d , q ) ,

.
AR ( p ) .

EViews.

:
yt = 0 + 1 yt 1 + K + p yt p + t ,
(*)

33

t ~ WN (0, 2 ) . EViews
( ). , ,
y110,
.
EViews
ls y5 c y5(-1).
EViews,
Equation: UNTITLED Workfile: CP-2
View Procs Objects Print Name Freeze

Estimate Forecast Stats Resids

Dependent Variable: Y5
Method: Least Squares
Date: 08/13/03 Time: 17:18
Sample(adjusted): 2 1000
Included observations: 999 after adjusting endpoints
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
Y5(-1)

5.055556
-0.503753

0.096585
0.027201

52.34281
-18.51944

0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat

0.255953
0.255207
0.987497
972.2257
-1403.950
1.994097

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

3.363014
1.144243
2.814714
2.824538
342.9695
0.000000

.
,
:
Estimate / ;
Forecast
;
Stats ;
Resids
(Residual), (Actual) ,
(Fitted).
EViews ar(1), ar(2),, ar(p),

( ):

yt = 0 + ut ,
(**)

ut = 1ut 1 + K + put p + t .

10

y5 , 5,
, (-0,5).

34

, i i (*) (**)
. ,
, :
ls y5 c ar(1),
:
Dependent Variable: Y5
Method: Least Squares
Date: 08/13/03 Time: 19:24
Sample(adjusted): 2 1000
Included observations: 999 after adjusting endpoints
Convergence achieved after 3 iterations
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)

3.361958
-0.503753

0.020777
0.027201

161.8136
-18.51944

0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
Inverted AR Roots

0.255953
0.255207
0.987497
972.2257
-1403.950
1.994097

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

3.363014
1.144243
2.814714
2.824538
342.9695
0.000000

-.50


: , ,
, ,
.
, (**) :
, Convergence achieved after 3
iterations - 3 . ,
,
(Inverted AR Roots).
( )
EViews. ,
Object/New Object/Equation,
Equation Specification

Equation Specification
Dependent variable following by list of regressors including ARMA
and PDL terms, OR an explicit equation like Y=c(1)+c(2)*X
Y5 c y5(-1)

5
5
5
6
6
6

Estimation Settings
Method:

LS Least Squares (NLS and ARMA)

OK
6
Cancel

Sample:

1 1000

5
5
5
6
6
6

Options

35

(*). , Method
:
LS ;
TSLS ;
ARCH ;
GMM ;
BINARY (logit, probit, extreme value);
ORDERED ;
CENSORED , ;
COUNT , ,
.

:
.
EViews , Quick/Estimate
Equation EViews. ,
, ,
.
,
ARMA( p, q ) , .
1.
, ,
, .
2.
, , ,
yt = c + a1 yt 1 + a2 yt 2 + t ,

, .. ,
,
ls y c y(-1) y(-2)
(
ls y c ar(1) ar(2)).
, ,
ls y c y(-2),

yt = c + a 2 y t 2 + t ,
,
.

.

36

3. ,

,
,
EViews.
, :
R 2 (R-squared).
R2 = 1

( y ) ( y )

yt , -
.

yt ,
.
,
0 1.
(, )
.
2
Radj
(Adjusted R-squared).

2
= 1 1 R2
Radj

)TT 1k ,

k .
R 2
,
.

,
.

,
,
.
s.e.regr. (S.E. of regression).
s.e.regr. =

T k
RSS (Sum squared resid).
,

.
2

RSS = SSR = = yt X t b ,

t =1

X t , b

37

l (Log likelihood).

T
l = 1 + log(2 ) + log
.
T
2
,
.
- DW (Durbin-Watson stat).
( )
.
T

DW =

(
t =2

t 1 )

t
t

2
T

T =1

(Mean dependent var S.D. dependent var).



:
T

=
t =1

yt
T

y =

t =1

( yt )2
T 1

AIC (Akaike info criterion).


l
k
AIC = 2 + 2 ,
T
T
l , k
. ,
,

, .
BIC SC (Schwarz criterion).
l k log T
BIC = SC = 2 +
.
T
T

, ,
. ,
,
.
F- (F-statistic). F- ,
,
, .. ,
, ,
, .

38

R2
F=

(k 1)

(1 R )
2

(T k )

k , ..
( ). EViews P F- (Prob(F-statistic)). P- ,
, ,
, ,
. ,
, .
4.
Residual Tests View .
:
Correlogram Q-statistics
( . 2.5);
Correlogram Squared Residuals
;
Histogram Normality Test
( . 2.3);
Serial Correlation LM Test
-. ,
p:
t = 1 t 1 + 2 t 2 + K + p t p + ut ,
u t ~ WN (0, u2 ). ,
:
H0 : 1 = K = p
:

H A : 12 + K + p2 > 0 .
TR 2 , T
, R 2

et = 0 + 1 x1 + K + k xk + 1et 1 + 2 et 2 + K + p et p + u t ,
et yt = 0 + 1 x1 + K + k xk + t . TR 2
2 ( p ) .

,
ARCH;
White Heteroskedastisity (no cross terms)
(
);
ARCH LM Test

39

White

Heteroskedastisity

(cross

terms)


).

q MA(q )

ma(1), ma(2), EViews. ,

yt = c + t + 1 t 1 + 2 t 2 + 3 t 3 ,

ls c ma(1) ma(2) ma(3).
,
ma(q),
.
ma3
Dependent Variable: MA3
Method: Least Squares
Date: 09/09/03 Time: 17:24
Sample(adjusted): 4 998
Included observations: 995 after adjusting endpoints
Convergence achieved after 30 iterations
Backcast: 1 3
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
MA(1)
MA(2)
MA(3)

1.028819
-1.383183
0.640902
-0.186233

0.017601
0.031186
0.049815
0.031133

58.45392
-44.35218
12.86569
-5.981893

0.0000
0.0000
0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
Inverted MA Roots

0.669620
0.668620
7.697083
58711.88
-3440.477
1.915246
.90

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
.24+.39i

1.022033
13.37097
6.923572
6.943281
669.5251
0.000000

.24 -.39i


. , Inverted
MA Roots ,

.

(p, q) ARMA( p, q )
EViews
, . ,


. ,
ARMA (2, 1), , ,

40

ls c ar(1) ar(2) ma(1)11.




,
, -:
Dependent Variable: ARMA21
Method: Least Squares
Date: 09/10/03 Time: 17:11
Sample(adjusted): 3 1000
Included observations: 998 after adjusting endpoints
Convergence achieved after 9 iterations
Backcast: 2
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
AR(2)
MA(1)

3.746622
1.495641
-0.894885
-0.524807

0.184022
0.016269
0.014565
0.032300

20.35962
91.93400
-61.44169
-16.24770

0.0000
0.0000
0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat

0.832956
0.832452
4.878129
23653.37
-2995.689
1.951439

Inverted AR Roots
Inverted MA Roots

.75+.58i
.52

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

3.728876
11.91745
6.011401
6.031063
1652.174
0.000000

.75 -.58i

, ,
, EViews
, ,

(
):

(1 L K
1

)(

Lp 1 L12 ut = t

)(

ut = 1 + 1 L + K + q Lq 1 + L12 t .

yt = c + u t ,
ut .
,

, :
ar(12)12 ma(12)
11

ar(1) ar(2) .
4: ar(4)
ma(4).

12

41

. ,
( s_ar),
EViews
ls s_ar c ar(1) ar(12).
,
:
Dependent Variable: S_AR
Method: Least Squares
Date: 07/15/03 Time: 14:34
Sample(adjusted): 14 1000
Included observations: 987 after adjusting endpoints
Convergence achieved after 13 iterations
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
SAR(12)

2.006013
0.574613
-0.620051

0.245655
0.026095
0.025378

8.165992
22.02035
-24.43280

0.0000
0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
Inverted AR Roots

0.728619
0.728068
5.318454
27833.38
-3048.447
1.963491
.93 -.25i
.57
-.25 -.93i

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
.93+.25i
.25 -.93i
-.68 -.68i

.68 -.68i
.25+.93i
-.68 -.68i

2.017917
10.19894
6.183277
6.198155
1320.952
0.000000
.68+.68i
-.25+.93i
-.93 -.25i

-.93+.25i

ls s_ar ar(1) sma(12).


,
Dependent Variable: S_AR
Method: Least Squares
Date: 07/15/03 Time: 14:36
Sample(adjusted): 2 1000
Included observations: 999 after adjusting endpoints
Convergence achieved after 13 iterations
Backcast: -10 1
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
MA(12)

1.994558
0.698247
-0.485910

0.302074
0.022740
0.027931

6.602889
30.70528
-17.39699

0.0000
0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
Inverted AR Roots
Inverted MA Roots

0.702925
0.702328
5.532175
30482.54
-3124.888
1.983751
.70
.94
.47 -.82i
-.47+.82i

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)
.82+.47i
.00 -.94i
-.82+.47i

.82 -.47i
-.00+.94i
-.82 -.47i

2.012653
10.13974
6.262037
6.276772
1178.342
0.000000
.47+.82i
-.47 -.82i
-.94

42


, ..

yt = c + ut ,
(1 a1 L )u t = (1 + L12 ) t .


. ,
, ..
yt = c + ut ,
(1 a1 L )(1 L12 )u t = t .
, ..
yt = c + ut ,

1 a1 L L12 ut = t ,
ls s_ar c ar(1) ar(12).
, ,
:
Dependent Variable: S_AR
Method: Least Squares
Date: 07/15/03 Time: 14:37
Sample(adjusted): 13 1000
Included observations: 988 after adjusting endpoints
Convergence achieved after 3 iterations
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
AR(12)

2.007593
0.475595
-0.513877

0.149227
0.018907
0.019092

13.45325
25.15399
-26.91633

0.0000
0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
Inverted AR Roots

0.772290
0.771827
4.870138
23362.47
-2964.534
1.962955

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

.97 -.24i
.28+.90i
-.64 -.66i

.97+.24i
.28 -.90i
-.64+.66i

2.023927
10.19552
6.007153
6.022019
1670.336
0.000000

.71 -.66i
-.21 -.91i
-.88 -.24i

.71+.66i
-.21+.91i
-.88+.24i


.
, ar(1) ar(12)
s_ar(-1) s_ar(-12)
, .
M ARIMA( p, d , q )

ARIMA( p, d , q )
L :

(1 a L K a L ) y
p

= c + 1 + 1 L + K + q Lq t ,
43

= 1 L . EViews ,
:
d(y)
, .. yt yt 1 = (1 L ) yt = yt ;
d(y, n, s)

,
..

(n, s)
n
s
n
s
(1 L ) 1 L yt = 1 L yt ;

dlog(y, n, s)
(n, s) , ..
(1 L )n 1 Ls log yt = n 1 Ls log yt .

ARIMA( p, d , q )
d, ..

d yt = c + a1 L + K + a p Lp d yt + 1 + 1 L + K + q Lq t .

, ARIMA(1, 1, 0)

(1 a1 L )yt = c + t ,

EViews :
ls d(y) c d(y(-1))

ls d(y) c ar(1).
:
Dependent Variable: D(Y)
Method: Least Squares
Date: 07/15/03 Time: 18:53
Sample(adjusted): 3 1000
Included observations: 998 after adjusting endpoints
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
D(Y(-1))

2.106930
0.473341

0.190197
0.027914

11.07765
16.95697

0.0000
0.0000

R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat

0.224020
0.223241
4.874090
23661.73
-2995.865
1.956180

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

3.992921
5.530322
6.007746
6.017577
287.5388
0.000000

:
Dependent Variable: D(Y)
Method: Least Squares
Date: 07/15/03 Time: 18:57
Sample(adjusted): 3 1000
Included observations: 998 after adjusting endpoints
Convergence achieved after 3 iterations
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)

4.000559
0.473341

0.292955
0.027914

13.65589
16.95697

0.0000
0.0000

R-squared

0.224020

Mean dependent var

3.992921

44

Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood
Durbin-Watson stat
Inverted AR Roots

0.223241
4.874090
23661.73
-2995.865
1.956180

S.D. dependent var


Akaike info criterion
Schwarz criterion
F-statistic
Prob(F-statistic)

5.530322
6.007746
6.017577
287.5388
0.000000

.47

ARIMA( p, d , q ) ,

: ma(1), ma(2) ..

1.

wn, y1-y5
ARMA( p, q ) p, q 2 .

2.


.
.

.

3.

2.7. ( -)

EViews 4.1 :
- (Augmented Dickey-Fuller Test),
-13 (Phillips-Perron Test), --
(Dickey-Fuller GLS Test & Elliot-Rothenberg-Stock Point-Optimal Test),
--- (Kwiatkowski-Phillips-Shmidt-Shin Test
KPSS Tests) - (Ng-Perron Test).
-.

-
View/Unit Root Test
Unit Root Test

Test type
Augmented Dickey-Fuller
Test for unit root in
Level
1st difference
2d difference
Include in test equation

13

Lag length
Automatic selection
Schwartz Info criterion

Maximum lags:
User specified 4

- - EViews 3.1.

45

Intercept
Trend and intercept
None
OK

Cancel

:

(Level), (1st difference),
(2d difference);

(Intercept), (Trend and intercept),
(None);
(Lag lendgh).
,
, (Automatic selection)
( 14 (Schwartz Info criterion))15,
User specified ( ,
, 4 ).
, OK.
-
( ):
Series: Y1 Workfile: CP-2
View Procs Object Print Name Freeze
Sample Genr Sheet Stats Ident LineBar
Augmented Dickey-Fuller Unit Root Test on Y1
Null Hypothesis: Y1 has a unit root
Exogenous: None
Lag Length: 0 (Automatic based on SIC, MAXLAG=21)
Augmented Dickey-Fuller test statistic
Test critical values:
1% level
5% level
10% level

t-Statistic

Prob.*

-17.08816
-2.567279
-1.941140
-1.616486

0.0000

*MacKinnon (1996) one-sided p-values.


Augmented Dickey-Fuller Test Equation
Dependent Variable: D(Y1)
Method: Least Squares
Date: 07/25/03 Time: 18:40
Sample (adjusted): 2 1000
Included observations: 999 after adjusting
endpoints
Variable

Coefficient

Std.
Error

t-Statistic

Prob.

14


(Akaike) - (Hannan-Quinn),
, -.
15
EViews 3.1

46

Y1(-1)
R-squared
Adjusted R-squared
S.E. of regression
Sum squared resid
Log likelihood

-0.451460 0.02641
9
0.226358
0.226358
1.054303
1109.331
-1469.846

-17.08816

Mean dependent var


S.D. dependent var
Akaike info criterion
Schwarz criterion
Durbin-Watson stat

0.0000
0.002017
1.198658
2.944637
2.949549
2.018780 6

,
(Y1), (CP-2).
:
View Procs Object Print Name Freeze

Sample Genr Sheet Stats Ident LineBar

Augmented Dickey-Fuller Unit Root Test on Y1 ,


- Y1.
,
(Null Hypothesis: Y1 has a unit root) ,
, (Exogenous: None),

(Lag Length: 0 (Automatic based on SIC, MAXLAG=21)), ..
:
Y 1t = Y 1t 1 + et ,
:

H0 : = 0 .
- (Augmented
Dickey-Fuller test statistic, 0 ) P-,
1, 5 10%- :
Test critical values:

1% level
5% level
10% level

-2.567279
-1.941140
-1.616486


EViews,
.

1. DEMO.wf :
y6=0,99*y6(-1)+wn
y7=y7(-1)+wn
y8=0,7+y8(-1)+wn
2. .
.
3. -,
y2, y6-y8.

47

4. -:

? ,
?
2.8. ARMA(p, q)

EViews ARMA(p, q)
16
(IND1).
1998 . 2002 .
(, .)
Forecast
2002 . 2003 . ( ,
1998 . 2003
.):
Forecast

Forecast of IND1

Method

Series name:
Forecast name:

Dynamic
Static

Ind1f

Structural

S.E. (optional):

(ignore ARMA)

G
G
A
R
C
H
GA
AR
RC
CH
H (((oooppptttiin
innaaall)l)):::
Output
Forecast sample

2002:10 2003:04

Insert actual for out-of-sample


OK

Do graph
Forecast
evaluation

Cancel

:
(Forecast name), (S.E. (optional)),
, (Forecast sample),
(Method: Dynamic, Static Structural) ,
(Output: Do graph Forecast evaluation).
:
. ,

h . ,
,
.
2 yt 1
16

1993 .

48

,
, yt 2 .
3 ,

,

, , :

80

Forecast: IND1F
Actual: IND1
Forecast sample: 2002:10 2003:04
Included observations: 7

78
76
74

Root Mean Squared Error


Mean Absolute Error
Mean Abs. Percent Error
Theil Inequality Coefficient
Bias Proportion
Variance Proportion
Covariance Proportion

72
70
68
66

2.137304
1.886911
2.654802
0.014775
0.092529
0.492510
0.414961

64
02:11

02:12

03:01

03:02

03:03

03:04

IND1F

.
, ..
, 1 T,
ARMA(p, q) ~
yT +1 .
7 ,
:

80

Forecast: IND1F
Actual: IND1
Forecast sample: 2002:10 2003:04
Included observations: 7

76

Root Mean Squared Error


Mean Absolute Error
Mean Abs. Percent Error
Theil Inequality Coefficient
Bias Proportion
Variance Proportion
Covariance Proportion

72

68

2.462476
1.944386
2.717071
0.017053
0.025005
0.175971
0.799024

64
02:11

02:12

03:01

03:02

03:03

03:04

IND1F

49

,
, ,
:
Root Mean Squared Error :
T +h

RMSE =

( y t yt )2
h +1

t =T +1

h , y t
, yt ;
Mean Absolute Error

:
T +h

MAE =

t =T +1

Mean Abs. Percent Error

y t yt
;
h +1

( %):

MAPE =

T +h

t =T +1

Theil Inequality Coefficient

y t yt
yt
;
h +1

:
T +h

TIC =

( y t yt )2
h +1

t =T +1
2

T +h

y t
+

t =T +1 h + 1

T +h

yt

t =T +1 h + 1


:
Bias Proportion

y t

y
h

,
2
( y y )
tht
y ;

:
Variance Proportion

(s

sy )

( y t yt )2

h
s y s y
;
Covariance Proportion
:

50

2(1 r )s y s y

( y t yt )2

h
r y t yt .

1. wn, y2, y4-y8 1


100 ARIMA( p, d , q ) .
2. 101, , 110

3.
?

?
3.
3.1. (
)
:
Ctrl
. ,
,
Open/as Group.

. ,
, Label+/- Wide+-,
2.1.

Object/New object/Group EViews
group < > < >,
.

1. DEMO.wf
rw=rw(-1)+nrnd
rw1=rw1(-1)+nrnd
rw2=rw+nrnd
rw3=1+rw+nrnd
rw4=-1+0,5*@trend+rw+nrnd.
2. RW, .

51

3.2.
View
Graph Multiple Graph.
Graph,
, :

730
720
710
700
690
680
670
660
960

970

980
Y8

990

1000

Y9

, Multiple Graph,
:

52

730
725
720
715
710
705
700
695
960

970

980

990

1000

990

1000

Y8

710
700
690
680
670
660
960

970

980
Y9

2.3.

53

1. DEMO.wf
RW.
2.

,
RW.
3.3.
Procs
Make Vector Autoregression
VAR Specification
Basics

? X

Cointegration

VEC Restrictions

VAR Type

Endogenous Variables

ind1 oilex

Unrestricted VAR
Vector Error Correction

Lag Intervals for Endogenous


12
Estimation Sample

Exogenous Variables

1995:01 2003:04

OK

, .. ,
(Endogenous Variables),
(Exogenous Variables),
, .. (Lag
Intervals for Endogenous), ,
(Estimation Sample).

: IND1
17
, OILEX

( 1995 . 2003 .):
Vector Autoregression Estimates
Date: 07/22/03 Time: 19:57
Sample: 1995:01 2003:04
Included observations: 99
Standard errors in ( ) & t-statistics in [ ]
IND1(-1)

17

IND1

OILEX

0.835523
(0.10137)
[ 8.24255]

0.614035
(0.09675)
[ 6.34644]

1993 .

54

IND1(-2)

-0.098109
(0.10096)
[-0.97172]

-0.612509
(0.09637)
[-6.35589]

OILEX(-1)

-0.362132
(0.08288)
[-4.36909]

0.050454
(0.07911)
[ 0.63776]

OILEX(-2)

0.545722
(0.08279)
[ 6.59160]

0.981794
(0.07902)
[ 12.4244]

1.127190
(3.00032)
[ 0.37569]

-2.389568
(2.86374)
[-0.83442]

0.826194
0.818875
614.2458
2.542783
112.8962
-232.6551
4.753102
4.883360
62.45052
5.974760

0.917102
0.913611
559.5961
2.427032
262.7452
-227.9961
4.659922
4.790181
84.16891
8.257463

R-squared
Adj. R-squared
Sum sq. resids
S.E. equation
F-statistic
Log likelihood
Akaike AIC
Schwarz SC
Mean dependent
S.D. dependent

Determinant Residual Covariance


Log Likelihood (d.f. adjusted)
Akaike Information Criteria
Schwarz Criteria

33.14213
-458.8280
9.376560
9.637077


() t [], ,
. 4
,
:
Determinant Residual Covariance
:

= det 1 t t ,

T p t

T , p ;
Log Likelihood (d.f. adjusted)
,
:
T
,
l = k (1 + log 2 ) + log
2
(Akaike Information Criteria Schwarz
Criteria):
2l 2n
AIC = +
T T

2l n log T
,
BIC = +
T
T
n=k(d+pk) ,
, k , d
( ).

55


, . -,
Residual Tests View
, ,
.
, Lag Structure View
:


(AR Roots Table AR Roots Graph):
Roots of Characteristic Polynomial
Endogenous variables: IND1 OILEX
Exogenous variables: C
Lag specification: 1 2
Date: 07/24/03 Time: 19:07
Root

Modulus

1.020817
0.811875
-0.473357 - 0.251051i
-0.473357 + 0.251051i

1.020817
0.811875
0.535811
0.535811

Warning: At least one root outside the unit circle.


VAR does not satisfy the stability condition.

,
( ) ,
, ..
;
,

(Pairwise Granger Causality Tests).

(Lag Exclusion Tests). (. )
2

,
( Joint):
VAR Lag Exclusion Wald Tests
Date: 07/24/03 Time: 18:21
Sample: 1995:01 2003:04
Included observations: 99
Chi-squared test statistics for lag exclusion:
Numbers in [ ] are p-values
IND1

OILEX

Joint

Lag 1

68.47932
[ 1.33E-15]

71.34502
[ 3.33E-16]

113.9743
[ 0.000000]

Lag 2

54.47254
[ 1.48E-12]

154.9752
[ 0.000000]

170.7268
[ 0.000000]

df

56

,
;
,
(Lag
Length Criteria).
, (
13):
VAR Lag Order Selection Criteria
Endogenous variables: IND1 OILEX
Exogenous variables: C
Date: 07/24/03 Time: 18:53
Sample: 1995:01 2003:04
Included observations: 99
Lag

LogL

LR

FPE

AIC

SC

HQ

0
1
2
3
4
5
6
7
8
9
10
11
12
13
14

-617.0574
-508.0099
-453.6986
-441.4990
-431.2965
-410.5607
-397.0080
-386.4080
-382.2395
-371.9166
-338.1122
-322.3627
-312.2120
-284.6735
-280.7090

NA
211.5521
103.1914
22.69131
18.56853
36.90985
23.58158
18.02011
6.919574
16.72324
53.41095
24.25423
15.22602
40.20621*
5.629569

816.8005
99.93104
36.53920
31.02164
27.41672
19.63368
16.23819
14.25299
14.23500
12.57770
6.953008
5.519332
4.904419
3.080743*
3.103997

12.38115
10.28020
9.273973
9.109980
8.985931
8.651213
8.460160
8.328159
8.324791
8.198331
7.602243
7.367253
7.244240
6.773470*
6.774180

12.43325
10.43651
9.534490
9.474704
9.454861
9.224351
9.137505
9.109710
9.210549
9.188296
8.696415
8.565631
8.546825
8.180261*
8.285179

12.40223
10.34346
9.379409
9.257591
9.175715
8.883172
8.734294
8.644467
8.683273
8.598988
8.045074
7.852258
7.771419
7.342824*
7.385708

* indicates lag order selected by the criterion


LR: sequential modified LR test statistic (each test at 5% level)
FPE: Final prediction error
AIC: Akaike information criterion
SC: Schwarz information criterion
HQ: Hannan-Quinn information criterion

6 ,
,
*.
3.4.
, .. ,
xt yt ,
EViews : View
Granger Causality ,
: IND1
18
, OILEX

( 1995 . 2003 .). , , , (. . 3.1).

18

1993 .

57

View/Granger Causality .
,
X

Lag Specifaication
Lags to include
OK

13
Cancel

(
13). :
Pairwise Granger Causality Tests
Date: 07/18/03 Time: 18:13
Sample: 1995:01 2003:04
Lags: 13
Null Hypothesis:

Obs

F-Statistic

Probability

OILEX does not Granger Cause IND1


IND1 does not Granger Cause OILEX

100

7.53840
1.32699

3.6E-09
0.21759

: 5%-
, OILEX
IND1,
, IND1 OILEX.
, EViews
, ..
:
yt = 0 + 1 yt 1 + K + p yt p + 1 xt 1 + K + p xt p + t

xt = 0 + 1 xt 1 + K + p xt p + 1 yt 1 + K + p yt p + u t

,
1 = K = p = 0
.
, ,
, .
3.5.

,
,

. ,
, , ,

.

58

,
View/Cointegration Test... ,
. ,
,

:
Johansen Cointegration Test

Cointegration Test Specification:


Deterministic trend assumption of test

Exog variables

Assume no deterministic trend in data:


1) No intercept or trend in CE or test VAR
2) Intercept (no trend) in CE no intercept in VAR
Allow for linear deterministic trend in data:
3) Intercept (no trend) in CE and test VAR

Do not include C or Trend


Critical values may not be
valid with exogenous
variables

4) Intercept and trend in CE no trend in VAR


Allow for quadratic deterministic trend in data:
5) Intercept and trend in CE linear trend in VAR

Lag intervals
14

Summary:
6) Summarize all 5 sets of assumptions

Lag spec for differenced


endogenous
OK

LR

, EViews :
1) No intercept or trend in CE or test VAR
;
2) Intercept (no trend) in CE no intercept in VAR
, ,
;
3) Intercept (no trend) in CE and test VAR
;
4) Intercept and trend in CE no trend in VAR
,
;
5) Intercept and trend in CE linear trend in VAR
.
,
( ) ( )
;
( )
( ) ( ) (
); .

59

, ,
,
6) Summarize all 5 sets of
assumptions. ,
:
Date: 07/19/03 Time: 17:46
Sample: 1995:01 2003:05
Included observations: 100
Series: IND1 OILEX
Lags interval: 1 to 13
Data Trend:

None

None

Linear

Linear

Quadratic

Rank or
No. of CEs

No Intercept
No Trend

Intercept
No Trend

Intercept
No Trend

Intercept
Trend

Intercept
Trend

Selected (5% level) Number of Cointegrating Relations by Model (columns)


Trace
Max-Eig

0
0

0
0

0
0

0
0

0
0

-286.5035
-281.8382
-280.7090

-286.5035
-280.3765
-276.4083

-281.4094
-276.4220
-276.4083

6.810070
6.787529
6.808166

6.748187
6.728440*
6.808166

8.216862
8.324580
8.475475

8.207083
8.291543
8.475475

Log Likelihood by Rank (rows) and Model (columns)


0
1
2

-288.6771
-285.4032
-283.7022

-288.6771
-282.5282
-280.7090

Akaike Information Criteria by Rank (rows) and Model (columns)


0
1
2

6.813542
6.828063
6.874044

6.813542
6.790565
6.854180

6.810070
6.796763
6.854180

Schwarz Criteria by Rank (rows) and Model (columns)


0
1
2

8.168231*
8.286958
8.437146

8.168231*
8.275512
8.469385

8.216862
8.307762
8.469385

, , ,
:
, . ,
: ( trace
Trace max Max-Eig)
5%- .
, , , ,
:
Date: 07/22/03 Time: 19:24
Sample(adjusted): 1995:01 2003:04
Included observations: 100 after adjusting endpoints
Trend assumption: Linear deterministic trend
Series: IND1 OILEX
Lags interval (in first differences): 1 to 12
Unrestricted Cointegration Rank Test
Hypothesized
No. of CE(s)

Eigenvalue

Trace
Statistic

5 Percent
Critical Value

1 Percent
Critical Value

None
At most 1

0.087027
0.013317

10.44561
1.340697

15.41
3.76

20.04
6.65

*(**) denotes rejection of the hypothesis at the 5%(1%) level


Trace test indicates no cointegration at both 5% and 1% levels
Hypothesized
No. of CE(s)

Eigenvalue

Max-Eigen
Statistic

5 Percent
Critical Value

1 Percent
Critical Value

None
At most 1

0.087027
0.013317

9.104913
1.340697

14.07
3.76

18.63
6.65

60

*(**) denotes rejection of the hypothesis at the 5%(1%) level


Max-eigenvalue test indicates no cointegration at both 5% and 1% levels
Unrestricted Cointegrating Coefficients (normalized by b'*S11*b=I):
IND1
0.277611
-0.293166

OILEX
0.015040
0.451230

Unrestricted Adjustment Coefficients (alpha):


D(IND1)
D(OILEX)

-0.221737
0.168112

1 Cointegrating Equation(s):

0.088506
0.096102
Log likelihood

-285.3438

Normalized cointegrating coefficients (std.err. in parentheses)


IND1
OILEX
1.000000
0.054175
(0.43561)
Adjustment coefficients (std.err. in parentheses)
D(IND1)
-0.061557
(0.03391)
D(OILEX)
0.046670
(0.03211)


.
:
Unrestricted Cointegration Rank Test

trace = T log(1 i ) ,
i = r +1

H 0 : rank r , H A : r = n ,
n 19,
max = T log(1 r +1 )


H 0 : rank r ,

H A : rank r + 1 .

(Trace Statistic
Max-Eigen Statistic), 5%- 1%- . ,

: Trace test indicates no cointegration


at both 5% and 1% levels Max-eigenvalue test indicates no cointegration at both 5% and 1%
levels.

. , ,

.

19

EViews 10 ,
.. n 10 .

61

p
:
Yt = A1Yt 1 + A2Yt 1 K + ApYt p + U t .

,
(Error Correction Model):
Yt = B1Yt 1 + B2 Yt 1 K + B p Yt p +1 + U t ,
p

i =1

i= j

B1 = I + Ai , B j = Ai ,

j = 2, p . .. Yt , K , Yt p+1

, Yt 1 .
, B1
. r.

B1 = T ,
n r T r n .
T Yt 1 r , T
,
.
, Unrestricted
Cointegrating Coefficients (normalized by b'*S11*b=I) ,
T . , ,
, r
1 r n 1 (
1 Cointegrating Equation(s): Normalized cointegrating
coefficients (std.err. in parentheses)).
,
, ..
.
,
,
, , ..
, p,
1 p-1, ..

1 p1.
, ,
1,
0 0.
3.3-3.5

1. DEMO.wf RW1, RW2, RW3, RW4,


: rw rw1, rw
rw2, rw rw3, rw rw4.

62

2.
, -
,
.
3.
, ,
(. 3.1)?
?

63