![](https://csdnimg.cn/release/download_crawler_static/18460014/bg1.jpg)
第
36
卷第
5
期
2013
年
5
月
合月巴工业大学学报(自然科学版)
JOURNAL OF HEFEI UNIVERSITY OF TECHNOLOGY
Doi:
10.
3969/j.
issn.
1003-5060.
2013. 05. 026
Vol.
36
No.
5
May
2013
基于
HMM
的
VaR
风险度量及其实证分析
.
江金菊
1
'
吴燕飞
1
'
王杨
2
(1.
合肥工业大学数学学院,安徽合肥
230009;
2.
云南大学经济学院,云南昆明
650091)
摘
要:文章基于隐马尔科夫模型(
HMM
)提出了度量金融资产风险价值
CVaR
)的
HMM-ARMA-G
且
.CH
模型。首先对金融资产收益率序列建立正常状态和异常状态的隐马尔科夫模型,使用期望最大化算法估算
出模型中的未知参数,再利用
Viterbi
算法估算出收益率序列所对应的隐状态序列,根据隐状态序列把收益率
序列数据分成正常状态类序列和异常状态类序列
2
个大类,对
2
个状态类序列分别建立
ARMAGARCH
模
型来估算
VaR
。最后利用该模型和传统的
ARMA-GARCH
模型对上证企债指数进行了实证分析,采用
Ku
piec
失败频率检验法对
VaR
的准确性进行检验。实证结果表明,该模型的
VaR
计算方法具有较好的估计效
果,能够有效地降低
GARCH
模型高估波动持续性的现象。
关键词:隐马尔科夫模型;
VaR
风险价值;
ARMAGARCH
模型;
Kupiec
失败频率检验
中图分类号:
0211.
62
文献标志码:
A
文章编号:
1003
5060(
2013
)05-0632-05
VaR risk measurement based on HMM and its empiricail analysis
WANG
Jin-ju1,
WU
Yan-fei1,
WANG
Yang2
Cl.
School
of
Mathematics,
Hefei
University
of
Technology,
Hefei
230009,
China;
2.
School
of
Economics,
Yunnan
University,
Kun
ming
650091,
China)
Abstract:
The
HMM-ARMA-GARCH
model
to
measure
the
financial
assets
value
at
risk
CVaR)
is
presented
based
on
the
hidden
Markov
model(
HMM).
First,
the
hidden
Markov
models
of
the
finan-
cial
asset
return
sequence
under
normal
and
abnormal
states
are
set
up.
The
expectation
maximization
algorithm
is
used
to
estimate
the
unknown
parameters
of
the
model.
Then,
the
Viterbi
algorithm
is
used
to
estimate
the
corresponding
hidden
state
sequence
of
the
return
sequence.
According
to
the
hid-
den
state
sequence,
the
return
sequence
is
classified
into
two
categories,
i.
e.
the
normal
state
se-
quence
and
the
abnormal
state
sequence.
And
the
ARMA
GARCH
model
is
established
to
estimate
VaR
for
each
state
sequence
respectively.
Finally,
the
Shanghai
enterprise
debt
index
sequence
is
ana-
lyzed
by
using
the
presented
model
and
the
traditional
ARMA-GARCH
model
respectively.
The
accu
racy
of
the
VaR
is
tested
by
the
Kupiec
failure
frequency
method.
The
results
show
that
the
presented
model
has
a
good
estimate
effect,
and
reduce
effectively
the
problem
of
overestimating
the
fluctuation
persistence
by
-the
GARCH
model
.
Key
words:
hidden
Markov
model;
value
at
risk(VaR);
ARMA-GARCH
model;
Kupiec
failure
fre
-‘
ql!ency
test
VaR
模型是
1993
年
1
P.
Morgan
在考查衍
生品的基础上提出的→种风险测度方法,该方法
收稿日期:
2012-09-13
;修回日期:
2013-01-04
以严谨的概率统计理论为基础,简单清晰地描述
了市场风险的大小,得到了国际金融界的广泛认
基金项目:安徽省自然科学基金资助项目(
1208085MF91;11040606M03
刘教育部人文社会科学研究资助项目(
10YJA910005
刘合肥
工业大学研究生教学改革资助项目
CYJG2010Y24
)和中央高校基本科研业务费专项资助项目(
2012
日
GXJ0043)
作者简介:汪金菊(
1978
一),女,安徽南陵人,博士,合肥工业大学讲师,硕士生导师.
第
36
卷第
5
期
2013
年
5
月
合月巴工业大学学报(自然科学版)
JOURNAL OF HEFEI UNIVERSITY OF TECHNOLOGY
Doi:
10.
3969/j.
issn.
1003-5060.
2013. 05. 026
Vol.
36
No.
5
May
2013
基于
HMM
的
VaR
风险度量及其实证分析
.
江金菊
1
'
吴燕飞
1
'
王杨
2
(1.
合肥工业大学数学学院,安徽合肥
230009;
2.
云南大学经济学院,云南昆明
650091)
摘
要:文章基于隐马尔科夫模型(
HMM
)提出了度量金融资产风险价值
CVaR
)的
HMM-ARMA-G
且
.CH
模型。首先对金融资产收益率序列建立正常状态和异常状态的隐马尔科夫模型,使用期望最大化算法估算
出模型中的未知参数,再利用
Viterbi
算法估算出收益率序列所对应的隐状态序列,根据隐状态序列把收益率
序列数据分成正常状态类序列和异常状态类序列
2
个大类,对
2
个状态类序列分别建立
ARMAGARCH
模
型来估算
VaR
。最后利用该模型和传统的
ARMA-GARCH
模型对上证企债指数进行了实证分析,采用
Ku
piec
失败频率检验法对
VaR
的准确性进行检验。实证结果表明,该模型的
VaR
计算方法具有较好的估计效
果,能够有效地降低
GARCH
模型高估波动持续性的现象。
关键词:隐马尔科夫模型;
VaR
风险价值;
ARMAGARCH
模型;
Kupiec
失败频率检验
中图分类号:
0211.
62
文献标志码:
A
文章编号:
1003
5060(
2013
)05-0632-05
VaR risk measurement based on HMM and its empiricail analysis
WANG
Jin-ju1,
WU
Yan-fei1,
WANG
Yang2
Cl.
School
of
Mathematics,
Hefei
University
of
Technology,
Hefei
230009,
China;
2.
School
of
Economics,
Yunnan
University,
Kun
ming
650091,
China)
Abstract:
The
HMM-ARMA-GARCH
model
to
measure
the
financial
assets
value
at
risk
CVaR)
is
presented
based
on
the
hidden
Markov
model(
HMM).
First,
the
hidden
Markov
models
of
the
finan-
cial
asset
return
sequence
under
normal
and
abnormal
states
are
set
up.
The
expectation
maximization
algorithm
is
used
to
estimate
the
unknown
parameters
of
the
model.
Then,
the
Viterbi
algorithm
is
used
to
estimate
the
corresponding
hidden
state
sequence
of
the
return
sequence.
According
to
the
hid-
den
state
sequence,
the
return
sequence
is
classified
into
two
categories,
i.
e.
the
normal
state
se-
quence
and
the
abnormal
state
sequence.
And
the
ARMA
GARCH
model
is
established
to
estimate
VaR
for
each
state
sequence
respectively.
Finally,
the
Shanghai
enterprise
debt
index
sequence
is
ana-
lyzed
by
using
the
presented
model
and
the
traditional
ARMA-GARCH
model
respectively.
The
accu
racy
of
the
VaR
is
tested
by
the
Kupiec
failure
frequency
method.
The
results
show
that
the
presented
model
has
a
good
estimate
effect,
and
reduce
effectively
the
problem
of
overestimating
the
fluctuation
persistence
by
-the
GARCH
model
.
Key
words:
hidden
Markov
model;
value
at
risk(VaR);
ARMA-GARCH
model;
Kupiec
failure
fre
-‘
ql!ency
test
VaR
模型是
1993
年
1
P.
Morgan
在考查衍
生品的基础上提出的→种风险测度方法,该方法
收稿日期:
2012-09-13
;修回日期:
2013-01-04
以严谨的概率统计理论为基础,简单清晰地描述
了市场风险的大小,得到了国际金融界的广泛认
基金项目:安徽省自然科学基金资助项目(
1208085MF91;11040606M03
刘教育部人文社会科学研究资助项目(
10YJA910005
刘合肥
工业大学研究生教学改革资助项目
CYJG2010Y24
)和中央高校基本科研业务费专项资助项目(
2012
日
GXJ0043)
作者简介:汪金菊(
1978
一),女,安徽南陵人,博士,合肥工业大学讲师,硕士生导师.