Tải bản đầy đủ
DANH MỤC TÀI LIỆU THAM KHẢO

# DANH MỤC TÀI LIỆU THAM KHẢO

Tải bản đầy đủ

PHỤ LỤC
1. Mô hình AR(1)

Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)

0.012658
0.655979

0.002924
0.064626

4.328436
10.15032

0.0000
0.0000

Inverted AR Roots

.66

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

0.013064
0.015695
-6.004287
-5.962461
-5.987291
1.940997

U

0.425689
0.421557
0.011937
0.019806
425.3023
103.0290
0.000000

́H

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

Ế

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 15:26
Convergence achieved after 3 iterations

K

IN

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 15:47
Convergence achieved after 3 iterations

H

2. Mô hình AR(2)

t-Statistic

Prob.

0.013087
0.429499

0.002105
0.077027

6.218053
5.575938

0.0000
0.0000

̣C

Std. Error

̣I H

C
AR(2)

Coefficient

O

Variable

0.183872
0.177958
0.014205
0.027847
397.9366
31.09108
0.000000

Inverted AR Roots

.66

Đ
A

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

2.

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

0.013200
0.015667
-5.656238
-5.614214
-5.639161
1.128392

-.66

Mô hình MA(1)

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 15:53

55

Convergence achieved after 4 iterations
MA Backcast: 2003M01
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
MA(1)

0.013057
0.638564

0.001727
0.064182

7.560095
9.949256

0.0000
0.0000

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

0.359448
0.354873
0.012576
0.022141
420.9070
78.56157
0.000000

Inverted MA Roots

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

0.013127
0.015657
-5.900099
-5.858468
-5.883182
1.749109

-.64

U

Ế

4. Mô hình MA(2)

Coefficient

Std. Error

C
MA(2)

0.013054
0.325472

0.001618
0.080472

IN

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

0.0000
0.0001
0.013127
0.015657
-5.603643
-5.562012
-5.586726
1.008212

O

̣C

K

0.138404
0.132250
0.014585
0.029782
399.8587
22.48920
0.000005

8.068365
4.044531

Prob.

̣I H

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

t-Statistic

H

Variable

́H

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 15:59
Convergence achieved after 17 iterations
MA Backcast: 2002M12 2003M01

5. Mô hình ARIMA(1,1,1)

Đ
A

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 16:01
Convergence achieved after 10 iterations
MA Backcast: 2003M02
Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
MA(1)

0.012668
0.633234
0.038852

0.002860
0.101102
0.130021

4.429582
6.263328
0.298816

0.0000
0.0000
0.7655

R-squared
S.E. of regression
Sum squared resid

0.425885
0.417564
0.011978
0.019799

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion

0.013064
0.015695
-5.990444
-5.927705

56

Log likelihood
F-statistic
Prob(F-statistic)

425.3263
51.18494
0.000000

Inverted AR Roots
Inverted MA Roots

Hannan-Quinn criter.
Durbin-Watson stat

-5.964949
1.968496

.63
-.04

6. Mô hình ARIMA(1,1,2)

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
MA(2)

0.012701
0.704830
-0.117135

0.003015
0.070805
0.098670

4.212591
9.954574
-1.187130

0.0000
0.0000
0.2372

́H

.70
.34

0.013064
0.015695
-5.997501
-5.934762
-5.972006
2.020757

Inverted AR Roots
Inverted MA Roots

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

H

0.429922
0.421660
0.011936
0.019660
425.8238
52.03609
0.000000

IN

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

U

Variable

Ế

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 16:03
Convergence achieved after 7 iterations
MA Backcast: 2003M01 2003M02

-.34

K

7. Mô hình ARIMA(2,1,1)

Variable

Coefficient

Std. Error

t-Statistic

Prob.

Đ
A

̣I H

O

̣C

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 16:09
Convergence achieved after 5 iterations
MA Backcast: 2003M03

0.013020
0.397635
0.754276

0.002892
0.097821
0.069709

4.501920
4.064933
10.82037

0.0000
0.0001
0.0000

C
AR(2)
MA(1)

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)
Inverted AR Roots
Inverted MA Roots

0.444596
0.436488
0.011761
0.018951
424.8779
54.83365
0.000000
.63
-.75

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

0.013200
0.015667
-6.026828
-5.963792
-6.001212
2.122995

-.63

57

8. Mô hình ARIMA(2,1,2)

Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(2)
MA(2)

0.013117
0.502437
-0.089818

0.002205
0.165315
0.191674

5.948482
3.039271
-0.468601

0.0000
0.0028
0.6401

Inverted AR Roots
Inverted MA Roots

.71
.30

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

-.71
-.30

K

IN

H

9. Mô hình SARIMA(1,1,1)(1,1,0)12
Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 22:06
Convergence achieved after 9 iterations
MA Backcast: 2004M02

0.013200
0.015667
-5.644300
-5.581264
-5.618684
1.115872

U

0.185786
0.173899
0.014240
0.027781
398.1010
15.63017
0.000001

́H

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

Ế

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/22/14 Time: 16:12
Convergence achieved after 14 iterations
MA Backcast: 2003M02 2003M03

Std. Error

t-Statistic

Prob.

0.013098
0.693322
0.365639
0.060687

0.005487
0.089604
0.083697
0.123468

2.387217
7.737591
4.368582
0.491523

0.0185
0.0000
0.0000
0.6239

̣C

̣I H

C
AR(1)
SAR(12)
MA(1)

Coefficient

O

Variable

0.494365
0.482230
0.011406
0.016263

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion

0.013961
0.015852
-6.078805
-5.990129

Log likelihood
F-statistic
Prob(F-statistic)

396.0829
40.73800
0.000000

Hannan-Quinn criter.
Durbin-Watson stat

-6.042774
1.996363

Đ
A

R-squared
S.E. of regression
Sum squared resid

Inverted AR Roots

Inverted MA Roots

.92
.46-.80i
-.46-.80i
-.92
-.06

.80+.46i
.46+.80i
-.46+.80i

.80-.46i
.00+.92i
-.80-.46i

.69
-.00-.92i
-.80+.46i

10. Mô hình SARIMA(1,1,3)(1,1,0)12
58

Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
SAR(12)
MA(3)

0.012995
0.689380
0.379286
0.176593

0.006055
0.067859
0.083383
0.091774

2.146057
10.15896
4.548728
1.924217

0.0338
0.0000
0.0000
0.0566

Inverted MA Roots

.80-.46i
.46-.80i
-.46-.80i

.80+.46i
.00+.92i
-.80-.46i

.28-.49i

11. Mô hình SARIMA(1,1,5)(1,1,0)12

U

.69
.00-.92i
-.80+.46i

-.56

Std. Error

t-Statistic

Prob.

0.012776
0.717590
0.365297
0.163221

0.006420
0.064519
0.083402
0.091103

1.989897
11.12218
4.379968
1.791608

0.0488
0.0000
0.0000
0.0756

Đ
A

̣I H

C
AR(1)
SAR(12)
MA(5)

Coefficient

O

Variable

̣C

K

IN

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/23/14 Time: 04:44
Convergence achieved after 8 iterations
MA Backcast: 2003M10 2004M02

0.013961
0.015852
-6.099520
-6.010844
-6.063489
1.839846

́H

.92
.46+.80i
-.46+.80i
-.92
.28+.49i

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

Inverted AR Roots

0.504732
0.492845
0.011289
0.015929
397.4190
42.46281
0.000000

H

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

Ế

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/23/14 Time: 04:42
Convergence achieved after 8 iterations
MA Backcast: 2003M12 2004M02

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)
Inverted AR Roots

Inverted MA Roots

0.506355
0.494508
0.011270
0.015877
397.6308
42.73948
0.000000
.92
.46+.80i
-.46+.80i
-.92
.56-.41i
-.70

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

0.013961
0.015852
-6.102803
-6.014127
-6.066772
1.885060

.80+.46i
.46-.80i
-.46-.80i

.80-.46i
.00-.92i
-.80+.46i

.72
-.00+.92i
-.80-.46i

.56+.41i

-.22+.66i

-.22-.66i

59

12. SARIMA(1,1,12)(1,1,0)12

Variable

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
SAR(12)
MA(12)

0.017812
0.756899
0.945753
-0.907528

0.021823
0.059852
0.035779
0.026535

0.816185
12.64625
26.43334
-34.20133

0.4159
0.0000
0.0000
0.0000

́H

U

0.013961
0.015852
-6.356250
-6.267574
-6.320219
1.874383

.86+.50i
.50-.86i
-.50-.86i

.86-.50i
.00+1.00i
-.86+.50i

.76
-.00-1.00i
-.86-.50i

.86-.50i
.00+.99i
-.86+.50i

.86+.50i
-.00-.99i
-.86-.50i

.50-.86i
-.50+.86i
-.99

H

Inverted MA Roots

1.00
.50+.86i
-.50+.86i
-1.00
.99
.50+.86i
-.50-.86i

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

IN

Inverted AR Roots

0.616872
0.607677
0.009929
0.012322
413.9781
67.08720
0.000000

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

Ế

Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/23/14 Time: 04:47
Convergence achieved after 10 iterations
MA Backcast: 2003M03 2004M02

K

 Đồ thị biểu hiện giá trị thực và giá trị dự báo

̣C

.08

O

.06

̣I H

.04
.02
.00

Đ
A

.04

-.02
.02
.00
-.02
-.04
04

05

06

07

08

Residual

09

10
Actual

11

12

13

14

Fitted

60

 RMSE, MAE, MAPE của mô hình dự báo
3.6

Forecast: CPIGOCFF
Actual: CPIGOC
Forecast sample: 2014M01 2015M12
Included observations: 11

3.5
3.4
3.3

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

3.2
3.1
3.0
2.9

0.055845
0.049623
1.738942
0.009722
0.789575
0.178639
0.031786

2014M07

2015M01

± 2 S.E.

́H

CPIGOCFF

2015M07

U

2.7
2014M01

Ế

2.8

 Đồ thị chuỗi phần dư của mô hình
RESID03

H

.04

IN

.03

K

.02

O

.00

̣C

.01

̣I H

-.01

Đ
A

-.02
-.03

03

04

05

06

07

08

09

10

11

12

13

14

 Kiểm định tính dừng chuỗi phần dư
Null Hypothesis: RESID03 has a unit root
Exogenous: Constant
Lag Length: 0 (Automatic based on SIC, MAXLAG=12)

Augmented Dickey-Fuller test statistic

t-Statistic

Prob.*

-10.66947

0.0000

61

Test critical values:

1% level
5% level
10% level

-3.482035
-2.884109
-2.578884

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

Variable

Coefficient

Std. Error

t-Statistic

Prob.

RESID03(-1)
C

-0.952933
0.000647

0.089314
0.000871

-10.66947
0.742918

0.0000
0.4589
-0.000177
0.013485
-6.394808
-6.350245
-6.376702
1.952522

U

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

́H

0.474645
0.470475
0.009813
0.012132
411.2677
113.8376
0.000000

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)

Ế

Augmented Dickey-Fuller Test Equation
Dependent Variable: D(RESID03)
Method: Least Squares
Date: 12/24/14 Time: 06:05

H

* Kiểm định tính nhiễu trắng của chuỗi phần dư

K

̣C

.|.
*|.
.|.
.|.
.|*
*|.
.|.
.|.
*|.
.|.
.|.
.|.
.|.
*|.
*|.
.|*
.|*
.|.
.|.
*|.
.|.
.|.

̣I H

|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|

Đ
A

.|.
*|.
.|.
.|.
.|*
*|.
*|.
.|.
*|.
.|.
.|.
.|.
.|.
*|.
*|.
.|*
.|*
.|.
.|.
*|.
.|*
.|*

Partial Correlation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|

O

Autocorrelation

IN

Date: 12/24/14 Time: 06:06
Sample: 2004M03 2014M11
Included observations: 129
Q-statistic
for 3 ARMA term(s)

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22

AC

PAC

0.046
-0.092
0.041
-0.017
0.158
-0.159
-0.104
0.035
-0.080
0.024
0.007
-0.033
0.004
-0.081
-0.166
0.106
0.098
-0.007
-0.000
-0.083
0.138
0.099

0.046
-0.094
0.051
-0.031
0.171
-0.194
-0.045
-0.010
-0.071
0.010
0.041
-0.029
-0.029
-0.066
-0.201
0.119
0.084
0.009
0.013
-0.066
0.045
0.061

Q-Stat
0.2831
1.3979
1.6251
1.6657
5.0543
8.5256
10.035
10.210
11.121
11.203
11.210
11.372
11.374
12.332
16.434
18.106
19.543
19.551
19.551
20.626
23.596
25.156

Prob

0.197
0.080
0.036
0.040
0.069
0.085
0.130
0.190
0.251
0.329
0.339
0.172
0.154
0.145
0.190
0.241
0.243
0.169
0.155

62

*|.
.|.
*|.
*|.
.|.
.|.
*|.
.|.
.|.
.|.
.|*
.|.
.|*
.|.

|
|
|
|
|
|
|
|
|
|
|
|
|
|

.|.
.|.
.|.
*|.
*|.
.|.
*|.
.|*
.|.
.|.
.|.
.|.
.|.
.|.

|
|
|
|
|
|
|
|
|
|
|
|
|
|

23
24
25
26
27
28
29
30
31
32
33
34
35
36

-0.086
-0.019
-0.100
-0.114
-0.065
-0.020
-0.097
0.019
0.025
0.062
0.095
0.022
0.115
0.003

-0.031
-0.030
-0.064
-0.168
-0.103
0.038
-0.158
0.076
0.067
0.040
0.035
0.049
0.047
0.045

26.344
26.401
28.020
30.135
30.836
30.903
32.509
32.567
32.671
33.347
34.921
35.005
37.363
37.364

0.155
0.192
0.175
0.146
0.159
0.192
0.177
0.212
0.248
0.264
0.246
0.284
0.236
0.275

U

Ế

* Kiểm định tính tự tương quan của mô hình

0.605123
0.000000

Prob. F(1,124)
Prob. Chi-Square(1)

Coefficient

Std. Error

t-Statistic

Prob.

C
AR(1)
SAR(12)
MA(12)
RESID(-1)

0.000734
-0.040455
-0.000934
0.001118
0.092891

0.021878
0.079361
0.035856
0.026619
0.119413

0.033536
-0.509759
-0.026038
0.041983
0.777896

0.9733
0.6111
0.9793
0.9666
0.4381

̣I H

O

̣C

K

Variable

IN

H

Test Equation:
Dependent Variable: RESID
Method: Least Squares
Date: 12/26/14 Time: 04:54
Sample: 2004M03 2014M11
Included observations: 129
Presample missing value lagged residuals set to zero.

0.4381
NA

F-statistic
Obs*R-squared

́H

Breusch-Godfrey Serial Correlation LM Test:

Đ
A

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

-0.001214
-0.033511
0.009944
0.012263
414.2921
1.968704

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.

0.000761
0.009782
-6.345614
-6.234769
-6.300576

13. Mô hình SARIMA(4,1,2)(1,1,1)12
Dependent Variable: D(CPI)
Method: Least Squares
Date: 12/24/14 Time: 04:25
Convergence achieved after 10 iterations
MA Backcast: 2004M03 2004M05
Variable

Coefficient

Std. Error

t-Statistic

Prob.

63

C
AR(4)
SAR(12)
MA(2)
SMA(1)

0.013297
0.254571
0.383517
0.488345
0.845745

R-squared
S.E. of regression
Sum squared resid
Log likelihood
F-statistic
Prob(F-statistic)
Inverted AR Roots

0.006053
0.118845
0.086760
0.107138
0.062548

0.511823
0.495685
0.011369
0.015641
387.8445
31.71521
0.000000

2.196676
2.142037
4.420411
4.558071
13.52156

Mean dependent var
S.D. dependent var
Akaike info criterion
Schwarz criterion
Hannan-Quinn criter.
Durbin-Watson stat

.92
.46-.80i

.80+.46i
.46+.80i

0.0299
0.0342
0.0000
0.0000
0.0000
0.014103
0.016010
-6.076897
-5.964347
-6.031172
2.109470

.80-.46i
.00-.71i

.71
.00+.92i

U
́H

.04

.03
.02

H

.01
.00

IN

-.01

-.03

05

06

07

08

09

10

11

12

13

14

D(CPIGOC) Residuals

O

̣C

04

K

-.02

Đồ thị thể hiện giá trị dự báo, giá trị thực của mô hình
dự báo và phần dư của mô hình

Đ
A

̣I H

Ế

 Đồ thị chuỗi phần dư của mô hình

64

.08
.06
.04
.02
.00

.04

-.02
.02
.00
-.02
-.04
04

05

06

07

08

09

11

Actual

12

13

14

Fitted

Ế

Residual

10

U

Kiểm định tính nhiễu trắng của chuỗi phần dư

-0.060
0.022
0.078
-0.009
0.124
-0.092
-0.115
0.063
-0.129
-0.013
-0.039
-0.130
-0.036
-0.048
-0.119
0.098
0.066
-0.021
0.044
-0.086
0.099
0.046
-0.150
0.137
-0.162
-0.078
-0.017
-0.035
-0.123
-0.004

IN

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30

PAC

H

AC

K

|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|

̣C

.|.
.|.
.|*
.|.
.|*
*|.
*|.
.|.
*|.
.|.
.|.
*|.
*|.
.|.
*|.
.|*
.|*
.|.
.|.
*|.
.|.
.|.
*|.
.|*
*|.
*|.
*|.
.|.
*|.
.|.

̣I H

|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|

Đ
A

.|.
.|.
.|*
.|.
.|*
*|.
*|.
.|.
*|.
.|.
.|.
*|.
.|.
.|.
*|.
.|*
.|.
.|.
.|.
*|.
.|*
.|.
*|.
.|*
*|.
*|.
.|.
.|.
*|.
.|.

Partial Correlation

O

Autocorrelation

́H

Date: 12/24/14 Time: 04:35
Sample: 2004M06 2014M11
Included observations: 126
Q-statistic
for 4 ARMA term(s)

-0.060
0.019
0.081
0.000
0.121
-0.086
-0.133
0.034
-0.109
-0.024
-0.021
-0.101
-0.079
-0.027
-0.121
0.075
0.124
-0.027
0.004
-0.097
0.013
0.016
-0.116
0.087
-0.160
-0.130
-0.087
0.042
-0.177
0.035

Q-Stat

0.4679
0.5333
1.3360
1.3461
3.3879
4.5284
6.3086
6.8437
9.1537
9.1771
9.3932
11.770
11.951
12.287
14.350
15.765
16.403
16.466
16.762
17.876
19.379
19.707
23.239
26.194
30.360
31.352
31.400
31.600
34.125
34.128

Prob

0.066
0.104
0.098
0.144
0.103
0.164
0.226
0.162
0.216
0.266
0.214
0.202
0.228
0.286
0.333
0.331
0.307
0.350
0.227
0.159
0.085
0.089
0.113
0.137
0.105
0.132

65