基于条件异方差模型GARCH对某股票价格数据进行拟合

本文选取的数据是某股票价格在28/9/2000-26/7/2017的收盘价格数据进行研究

使用了SAS系统的autoreg过程来进行条件异方差模型GARCH建模和条件异方差检验。

（1）首先建立数据集a

data a;
input x@@;
t=_n_;
cards;
111.94
102.5
104
95
99.5
97.75
92.75
87.19
96.75
100.44
103.94
106.56
113.5
111.31
106.25
112.25
110.19
110
101.62
103.75
100.25
98.81
103.31
108
114.06
114.87
112.62
113.5
108.75
115.81
119
114.12
114.87
111.25
113.81
116.25
115.2
110.37
113.25
114.62
130.31
122
128
125.69
122.87
122
122.25
125.75
121.31
114.25
116.25
121.56
125
123
134.94
144.19
141.31
138.69
139.56
139.13
135.94
125.75
135.81
133.31
127.31
130.38
125.19
131.75
125
119.44
109.25
113.81
111.87
123.87
126.87
121.12
118.87
120.5
128.31
121.31
126.75
124.06
124.31
117.87
115.06
110.69
113.12
110.12
105.44
99.31
102.81
107.62
101
105.69
101.37
100.75
94
89.94
85.81
87.69
87.27
86.37
87.56
84
89.12
92.56
91.31
92.87
96.56
94.81
95.75
91.19
94.5
90.44
92.37
91.19
96.62
101.25
55.63
53.75
51.69
54.13
51.75
54.44
51.25
52.38
53.31
51.63
51.81
54.44
57.13
56.94
58.88
56.5
57.69
58.31
57.25
52.69
55.13
53.56
48.69
50.06
50.06
52
48.31
50.81
49.31
47.25
48
47.38
47.94
46.75
47.5
47.56
47.69
47.06
46.69
48.5
51.44
50
50.5
51.69
54.31
56.11
56.81
58.06
59.19
59.5
60.94
63.44
62.44
58.44
62
58.88
58.44
57.75
58
56.86
55.23
60.66
59.94
61.05
56.69
52.19
53.5
51.44
48.94
53.5
25.75
24.25
22.31
23.62
22.06
22.19
21.75
20.87
19.62
20
22.06
21.5
20.12
20.12
18.94
19.5
20.37
18.87
18.5
18.5
18.56
19.31
19.56
20.5
22.31
22.25
21.44
21.31
20.06
20.19
19.06
19.37
20.25
19.87
19
18.5
18.94
18.81
18.5
19.31
18.69
18.03
17.56
16.5
17.06
16.69
17
14.31
14.31
15.06
15.19
15.38
15
14.44
14.06
14.25
14
14.38
14.06
15
14.69
14.81
14.81
14.88
14.88
16.37
17.06
16.37
16.56
17.19
16.56
18
17.19
17.12
16.81
18.69
19.5
19.25
20.5
20.5
19.94
19.56
21.69
21.75
21.62
21.12
20.62
20.19
21.12
20.75
20.75
19.12
19.69
19.12
19.5
20.06
19
18.31
18.87
18.81
18.81
19.5
19.37
18.25
18.75
19.25
20.37
21.5
21.25
20.81
20.25
18.62
19.56
20.44
19.69
19.62
20.56
19.69
20.12
21.62
23
21.78
22.87
22.17
22.53
22.07
21.59
20.24
19.5
20.87
20.59
20.54
22.04
21.8
22.42
21.44
20.4
22.79
25.72
25.04
24.25
24.03
24.72
24.69
26.2
25.49
25.93
26.59
24.96
25.75
24.96
24.57
23.98
23
22.85
23.29
23.18
24.1
23.55
23.53
23.56
23.5
23.23
23.2
22.76
21.47
19.78
19.95
20.89
20.66
20.94
20.73
21.66
21.32
20.04
20.31
20.47
19.88
20.44
20.33
20.19
21.67
22.49
22.26
23.99
23.75
23.34
23.54
23.25
23.9
23.84
23.19
22.03
22.7
21.14
22.54
24.36
24.85
23.96
25.1
20.79
19.96
19.98
19.54
19.09
18.47
18.59
18.96
18.93
18.79
19.06
19.82
19.5
19.13
19.25
18.9
19.05
19.02
19.09
18.73
18.44
18.65
18.07
18.12
17.92
18.21
17.81
18.57
18.92
18.4
17.83
17.83
18.55
18.25
18.55
17.72
17.28
17.37
16.99
16.28
17.02
15.68
15.73
16.45
15.54
15.15
15.51
15.51
15.54
15.05
14.98
15.88
16.14
16.2
16
16.82
17.74
18.01
17.99
18.01
16.99
18
18.3
19.02
18.14
18.95
19.19
18.67
17.63
17.6
17.56
18.59
18.57
19.07
19.57
19.59
18.71
18.71
18.75
19.37
19.61
19.45
18.97
20
19.53
19.68
19.84
21.37
21
20.53
20.42
21.3
21.05
22.4
23.76
22.78
22.54
22.54
21.78
21.49
21
20.39
20.62
21.01
21.62
20.67
21
21.36
21.49
22.07
22.43
21.9
23.3
23.58
23.69
22.9
22.61
21.65
21.23
21.05
21.15
21.7
20.78
22.48
22.17
21.82
23.02
23.21
23.25
23.27
23.07
24.09
24.72
24.41
25.35
25.45
24.67
24.3
24.03
24.98
24.71
25.01
24.6
23.9
22.62
23.13
21.5
22.74
23.81
23.67
21.96
21.7
23.45
24.29
23.53
24.07
24.38
24.66
25.06
24.72
24.49
24.43
24.95
24.74
24.85
24.92
24.27
24.09
23.35
23.46
23.47
23.67
24.46
24.07
23.75
24.9
24.74
24.56
24.1
24.66
24.86
25.06
25
25.74
26.11
25.41
24.98
24.53
24.25
23.77
24.12
23.01
23.96
24.27
23.98
23.69
23.51
22.65
22.47
24.37
24.19
23.32
23.94
25.61
25.28
25.21
25.01
24.74
23.46
24.32
25.18
24.15
23.98
23.98
24.2
23.3
22.91
22.78
22.72
22.16
21.4
21.48
20.46
20.09
19.54
20.1
20.54
20.15
17.12
17.11
16.85
17.27
17.14
16.55
17.06
17.72
17.06
16.94
17.55
18.74
18.01
17.53
17.32
18.3
17.5
18.23
17.86
15.63
14.99
14.96
14.92
14.47
15.2
14.36
14.34
15.02
15.43
15.26
14.8
14.45
13.99
14.74
15.03
15.3
15
14.99
14.59
15.17
15.61
15.81
15.98
15.91
16.12
15.97
15.73
15.53
14.85
14.7
14.7
14.75
14.05
14.48
14.18
14.38
14.37
14.33
14.29
14.14
14.17
14.5
14.8
15.02
14.58
14.87
14.85
14.64
14.93
14.7
14.72
14.5
14.51
14.17
14.3
14.03
13.77
13.69
13.6
14.12
14.51
14.77
15.16
14.56
14.11
14.34
14.56
14.7
14.88
14.69
15.42
15.61
15.44
15.98
16.07
16.36
16.89
16.9
17.22
16
15.84
15.16
15.64
15.59
16.3
15.95
15.65
15.27
15.53
16.35
16.01
15.97
15.41
15.72
15.5
15.18
15.16
14.97
14.63
14.95
14.75
15.28
15.49
15.19
14.79
14.85
15.08
14.57
14.2
14.14
14.49
14.36
14.4
14.06
14.07
14.33
14.8
14.9
14.9
14.85
14.55
14.68
14.72
14.63
14.61
14.43
14.62
14.1
14.02
13.88
14.17
13.8
14.13
14.58
14.58
14.32
14.36
14.66
14.6
14.45
14.43
14.15
14.35
14.35
14.39
14.54
14.67
15.27
14.85
14.77
15
14.74
15.02
14.5
14.86
15.01
14.65
14.56
14.62
14.56
14.53
14.37
14.23
14.22
14.72
14.78
15.01
15
14.95
14.91
15
14.37
14.55
14.41
14.49
14.57
14.14
14.16
14.6
14.46
14.41
14.49
14.45
14.19
14.37
13.2
13.58
13.39
13.24
13.12
13.14
13.51
13.58
13.44
13.35
13.86
14.06
14.22
14.36
14.45
16.09
17.5
17.65
18
18.3
18.56
18.67
18.55
18.73
18.8
18.1
17.79
17.85
18.24
18.32
18.88
18.28
18.1
17.95
17.45
17.31
17.6
17.64
17.15
16.79
17.18
17.45
17.77
17.42
18.27
18.19
19.12
19.14
19.2
19.06
18.78
19.09
19.29
18.73
19.06
19.09
19.27
19.13
19.87
20.4
19.89
19.58
19.85
19.9
19.61
19.87
20.9
20.859
20.61
20.8
20.79
20.51
21.54
20.99
20.72
20.28
21.08
20.73
21.21
20.38
19.63
19.93
19.64
19.66
19.7
20.18
19.97
19.71
20.34
20.32
21.01
21.68
20.88
20.86
21.05
21.48
22.19
22.61
22.85
22.95
22.83
22.5
22.74
22.37
22.18
22.56
23.1
22.21
22.36
22.12
22.88
22.58
22.08
22.43
21.32
20.43
20.69
21.3
20.72
20.79
20.57
21.69
22.29
23.22
23.06
23.45
23.68
24.35
24.55
24.82
23.25
22.75
23.22
23.18
22.76
22.99
22.6
22.6
23.72
23.69
23.09
22.89
23.15
22.91
23.03
23.12
22.5
21.9
21.54
22.33
22.42
21.46
21.13
20.41
20.4199
20.375
20.28
21.15
20.68
20.72
20.91
21.71
21.54
21.03
21.15
20.85
21.05
20.45
20.38
21.21
20.89
20.17
20.12
19.88
20.04
19.7
19.85
19.81
20.41
20.78
21.15
21.28
21.37
21.28
22.17
22.09
22.59
23.36
23
23.73
24.12
24.2
22.85
22.72
22.73
22.61
22.18
22.56
23.01
23.07
22.52
22.68
22.56
22.32
22.26
21.79
22.42
22.71
22.667
22.98
23.8
23.73
23
23.16
23.26
22.47
22.4
22.19
22.36
22.81
23.04
23.92
24.02
23.81
23.92
25.16
26.74
26
27.1
27.68
27.15
27.56
26.45
25.82
26.19
25.67
25.86
25.86
25.29
25.5
26.87
27.04
27.91
27.92
27.04
27.11
27.5
28.32
27.83
27.31
27.53
28.04
26.93
26.64
29.3
29.18
28.35
27.73
27.73
27.78
27.7
27.13
26.94
26.45
26.77
25.78
26.07
26.14
26.65
26.58
26.67
26.28
27.14
27.3
27.19
27.06
26.64
27.06
26.47
26.71
27.11
27.34
28.41
28.51
28.17
28.06
28.06
28.92
28.4
28.78
29.81
30.35
30.2
30.74
30.12
30.69
32.74
32.81
32.91
32.33
33
33.7
33.18
33.7
32.49
32.5
32.54
32.3
31.08
30.95
30.39
30.14
30.03
29.14
29.22
29.58
32.93
32.2
31.97
32.2
31.62
31.68
30.7
31.26
32.43
32.27
32.64
32.34
31.58
31.29
31.79
31.39
29.78
30.3
31.52
31.01
30.37
30.84
30.78
30.87
31.74
30.71
30.8
31.08
31.95
33.05
34.66
34.35
34.12
34.49
35.86
35.66
35.23
35.76
36.35
35.7
35.87
35.59
35.49
35.2
36.35
37.14
37.71
38.01
36.92
37.27
37.29
37.53
38.04
38.68
38.75
38.67
38.79
39.37
40.64
39.62
39.06
38.59
38.29
39.75
44.98
45.5
47.75
47.42
47.47
47.94
47.41
47.55
47.97
50.3
52.19
52.4
52.45
53.5
55.31
54.45
54.72
54.38
54.05
54.75
55.3
55.5
55.24
54.935
54.9
55.39
55.17
61.35
61.27
64.05
64.55
68.44
67.05
67.79
65.21
62.68
65.78
62.89
63.28
63.99
65.15
64.91
65.29
65.26
66.596
64.99
62.72
63.69
63.75
64.01
63.16
64.18
64.44
64.8
64.4
63.29
63.94
64.5
64.55
69.25
68.96
64.56
65.46
69.8
70.2
70.65
69.88
70.46
70.49
70.76
72.05
72.25
72.64
73.98
76.9
77.53
79.63
77.81
78.84
78.94
80.9
78.74
78.36
81.21
84.63
88.41
90.13
87.81
86.81
85.29
88.23
88.93
88.99
44.86
44.5
44.121
41.79
42.81
42.75
40.53
39.35
39.83
40.27
40.32
40.96
41.18
42.25
42.96
43.7
42.83
42.55
42.5
42.53
41.75
42.8
41.67
40.89
41.09
41.89
42.33
43.56
43.74
41.92
42.66
41.04
37.26
35.35
35.62
37.09
35.51
37.14
35.5
36.98
36.19
35.95
35.54
36.06
36.43
36.21
37.15
36.68
37.24
36.97
36.42
35.61
34.13
34.77
35.55
35.36
35.84
37.55
37.55
39.76
39.7
39.78
40.74
40.56
39.7593
40.3
40.04
38.24
37.92
36.54
36.92
37.65
35.81
35.9
36
37.13
37.98
38.31
37.61
37.86
38.55
38.89
37.76
37.1
37.31
36.37
36.81
36.5
37.98
37.39
37.63
38.25
38.1
38.24
38.35
40.75
41.55
41.49
43.19
43.63
43.29
44
43.81
43.63
43.99
43.8
42.65
42.75
43.19
43.22
42.71
42.99
42.65
43.82
43.38
44
46.1
47.68
46.25
47.15
46.3
45.83
45.87
45.74
45.77
46.06
45.74
45.84
46.57
46.89
46.26
46.22
48.8
48.68
49.78
51.31
51.4
50.82
49.61
49.87
51.21
52.64
53.19
52.11
51.9
53.2
53.84
53.44
51.08
52.34
53.61
54.44
53.75
52.7801
51.7
51.3
50.37
51.59
49.25
53.74
54
53.44
52.21
54.94
56.14
55.66
56.79
56.1
57.03
55.41
54.47
57.59
57.5
59.95
61.8499
61.15
60.23
59.9
60.11
61.18
61.54
61.45
62.28
64.95
64.52
64.56
64.96
66.52
67.11
69.34
69.66
68.1
67.82
71.6001
72.63
71.82
74.05
73.95
74.08
74.3305
74.91
74.98
72.01
72.18
71.11
71.38
72.11
73.5
74.02
73.35
74.23
73.57
71.45
71.89
74.75
74.97
74.38
76.3
76.05
80.86
83.9
84.291
85.5899
84.7101
82.4901
79.035
76.089
77.67
76.04
74.2
72.33
72.03
75
75.51
75.42
72.1
71.847
67.3
67.6
68.81
64.95
67.31
64.71
67.645
69.22
70.57
70.29
69.08
71.32
71.75
71.46
70.99
68.49
69.1
69.61
67.72
65.48
66.31
65.66
63.93
63.19
65.68
67.32
66.23
64.31
64.66
63.99
61.81
61.67
60.16
59.96
59.51
58.71
62.33
62.75
62.72
62.65
61.17
67.21
71.24
69.79
68.67
67.99
66.71
66.469
64.811
66.22
65.65
67.63
67.04
65.75
66.17
68.15
...
597.51
604.59
604.71
606.31
607.27
614.13
625.63
624.01
635.38
633
628.65
637.54
644.82
647.35
645.57

;

(2)绘制股票价格走势的时序图

proc gplot data=a;
plot x*t=1;
symbol1 c=black i=join v=star;
run;

运行得到：

因为股价走势是随时间走势趋势向上的，本文考虑用股票价格变量x对时间变量t做回归，提取序列中的趋势信息。

(3)股票价格变量x对时间变量t做回归,并进行DW检验和条件异方差检验

proc autoreg data=a;
model x=t/nlag=5 dwprob archtest;
run;

运行得到：

异方差检验显示提取趋势信息后该残差序列为显著的异方差序列，

显示残差序列具有显著正相关。

（3）残差序列用2阶自回归，AR(2)-GARCH(1,2)模型来模拟该股票价格数据。

proc autoreg data=a;
model x=t/nlag=2 noint garch=(p=1,q=2);
output out=out p=p residual=residual lcl=lcl ucl=ucl cev=cev;
run;

运行得到：

模型的总R方为0.9996,下面是模型的拟合参数。

再次考虑用AR(2)-GARCH(0,1)模型来模拟该股票价格数据。

（4）用AR(2)-GARCH(0,1)模型来模拟该股票价格数据。

proc autoreg data=a;
model x=t/nlag=2 noint garch=(q=1);
output out=out p=p residual=residual lcl=lcl ucl=ucl cev=cev;
run;
运行得到：

各阶参数估计显著

（5）绘制拟合效果图

data out;
set out;
l95=-1.96*sqrt(29.04145);
u95=1.96*sqrt(29.04145);
Lcl_GARCH=-1.96*sqrt(cev);
Ucl_GARCH=1.96*sqrt(cev);
proc gplot data=out;
plot residual*t=2 l95*t=3 u95*t=3 Lcl_GARCH*t=4 Ucl_GARCH*t=4/overlay;
plot x*t=2 lcl*t=3 ucl*t=3 /overlay;
symbol2 c=yellow i=join v=none ;
symbol3 c=red i=join v=none;
symbol4 c=blue i=join v=none;
run;

运行得到：

本文残差序列假定服从正态分布，但检验结果不符合，总体拟合效果较好。

对于错误的地方还请各位读者批评指正，相互学习相互进步。