模糊随机理论研究及生命年金应用
VIP免费
摘 要
随着我国人口年龄结构趋于老龄化,养老金制度的“三大支柱”尤其是社会
基本养老保险的偿付压力越加巨大。如何准确定价年金、管理年金,缓解养老金
偿付压力,是关乎国家经济健康良好发展的大事。
本文构建生命年金的定价模型是基于两个不确定因素,分别是贴现率(即利
率)的波动性和死亡危险率的变化。
传统的生命年金现值精算是基于固定利率下的研究,这显然不符合实际情况。
由于金融市场的利率存在波动性,学者们采用多种方法进行研究,如 Markov 链随
机利率、分形方法等。
利率的不确定性表现为模糊性,往往存在这样的术语,例如,未来
n
年的利率
预期为“比较高”或“比较低”。由于死亡率的随机性和利率的模糊性,年金现值
精算过程是不确定的过程。对于贴现率估计,本文介绍模糊贴现率估计方法,分
别是模糊平均收益率、模糊浮动利率、模糊利率的期限结构理论框架。其中,模
糊利率的期限结构理论框架比较接近本文需求,故采用模糊利率
(1 ) t
tt
vi
。
本文构建模糊随机变量年金定价模型,分别从离散和连续角度分析年金现值,
并给出案例分析,表明模糊随机变量相关特征量(均值、方差、分位数、分布函
数)比较完好地描述年金现值。
作为年金管理者需要提供充足的准备金以应对后续剩余的年金支付款项,给
出年金组合的现值和方差,以便评估年金组合的风险。为了精确估计风险成本的
大小,研究年金组合的分位数是很有必要的,而分位数可以根据分布函数得到。
然而,年金组合分布函数相关的
水平截集表达式比较难获得。为此,分别提出
了两种近似估计方法,即随机模拟方法和中心极限方法,并且分别对两种方法给
出案例分析。对于分位数逼近,两种方法得到较为接近的结果,中心极限方法的
计算速度快。对于分布函数逼近过程中,两种方法得到的结果收敛性较为接近。
最后,将生存模型的随机死亡危险率改进为模糊随机死亡危险率,构建剩余
寿命模糊随机模型。通过不同的模糊语言变量结合 Gompertz 死亡危险率函数,分
析不同模糊语言环境的剩余寿命模型。对于未来年金定价的研究而言,可以进行
扩展,如评估各种累积和折扣模型等。
总之,模糊随机理论应用于年金定价,既刻画了人们主观决策意愿,又不缺
失市场变动造成的随机性波动。对于年金的风险管理方面,做出完整的风险描述,
使得年金管理者对于未来支付项如何进行准备金评估得到很大帮助。
关键词:年金定价 模糊随机变量 生存模型 贴现率 死亡危险率
ABSTRACT
Along with the aging of population in China, “three pillars” of the pensions system,
especially, the payment of social basic endowment insurance become more and more
stressful. How to accurately price, manage annuities and alleviate the pressure of
pension payments is a big deal about the development of the national economy in good
health.
This paper constructs a life annuity pricing model based on two uncertain factors,
which, respectively, is the volatility of discount rate (also interest rate) and the change
risk of mortality.
The actuarial present value of traditional life annuities is based on the research of
fixed interest rate. Therefore, it is not in conformity with the actual situation. Because of
the volatility of interest rate in financial market, the scholars study the pricing model by
using a variety of methods, such as the Markov chain stochastic interest rate, fractal
method etc.
The uncertainty of the interest rate expresses as fuzziness and there often exists
such jargon. For example, the expected interest rate of future
n
years is “relatively
high” or “relatively low”. Due to the randomness of mortality and the fuzziness of
interest, pricing actuarial present value of annuities is a uncertain process. When
estimating the discount rate, this paper introduces the approach such as fuzzy average
yield rate, fuzzy floating interest rate, the framework of term structure theory of fuzzy
interest rate. Among all above, the framework of term structure theory of fuzzy interest
rate is closest to the needs of the research, so we adopt the fuzzy interest rate as
(1 ) t
tt
vi
.
Constructing the pricing model for fuzzy random variable of life annuities, this
paper analysis the present value of annuities from the perspectives of discrete and
continuous respectively and gives numerical examples, which is given to show that the
related characteristics of fuzzy random variables (such as mean value, variance and
quantiles, distribution function) perfectly describe the present value of annuities.
As a pension manager, we need to provide sufficient reserve fund to cope with the
rest payments of annuities by giving the present value and variance of annuity portfolio
to assess the risk of annuity portfolio. To estimate the cost size of risk accurately, it is
extremely necessary to research the quantiles of annuity portfolio, and the quantiles can
be obtained according to the cumulative distribution function. However, the related
-cut of cumulative distribution function of annuity portfolio is difficult to be obtained.
To this end, two approximate estimation methods are proposed respectively, namely the
method of stochastic simulation and central limit, and numerical examples are given for
the two methods respectively. When approximating to quantiles, two methods both
obtain good results and the method of central limit’s calculating speed is fast than the
other. For the process of approximating cumulative distribution function, stochastic
simulation method has good convergence results.
Finally, while building the fuzzy stochastic model of future lifetime, the survival
model for the random risk of mortality is improved to a fuzzy random one. Considering
different fuzzy language variables, we analysis future lifetime model for different fuzzy
language environment by combining Gompertz mortality function. For the following
study of annuity pricing, it can be extended as the evaluation of various accumulation
and discount model.
In short, fuzzy random theory applied to pricing annuities not only describes human
subjective decision and willingness, but also full of market volatility caused by the
random fluctuations. For the aspect of annuity risk management, a complete description
of the risk makes pension managers prepare reserve fund to deal with future payment
items for annuities easily.
Key Words: Pricing Annuities, Fuzzy Random Variable, Lifetime
Model, Discount Rate, Instantaneous Force of Mortality
目 录
中文摘要
ABSTRACT
第一章 绪 论 .................................................................................................................. 1
1.1 选题背景及意义 ................................................................................................ 1
1.2 国内外研究现状 ................................................................................................ 2
1.3 研究方法及内容框架 ........................................................................................ 4
1.3.1 研究方法 ................................................................................................. 4
1.3.2 内容框架 ................................................................................................. 4
第二章 基本理论 ............................................................................................................ 6
2.1 生命年金的知识和基本原理 ............................................................................ 6
2.1.1 生命年金简介 ......................................................................................... 6
2.1.2 连续型生命年金 ..................................................................................... 7
2.1.3 离散型生命年金 ..................................................................................... 8
2.2 生存分析基本函数和生存模型 ........................................................................ 9
2.2.1 生存函数 ................................................................................................. 9
2.2.2 危险率函数 ........................................................................................... 10
2.3 模糊随机变量 ................................................................................................... 11
2.3.1 模糊随机变量定义 ................................................................................ 11
2.3.2 离散型模糊随机变量 ........................................................................... 12
2.3.3 连续型模糊随机变量 ........................................................................... 14
2.4 本章小结 .......................................................................................................... 15
第三章 生命年金定价 .................................................................................................. 16
3.1 模糊贴现率估计 .............................................................................................. 16
3.1.1 平均收益率 ........................................................................................... 16
3.1.2 浮动利率 ............................................................................................... 16
3.1.3 利率的模糊期限结构 ........................................................................... 17
3.2 离散型生命年金定价 ...................................................................................... 17
3.2.1 离散型生命年金现值估计 ................................................................... 18
3.2.2 案例分析 ............................................................................................... 20
3.3 连续型生命年金定价 ...................................................................................... 22
3.3.1 连续型生命年金现值估计 ................................................................... 23
3.3.2 案例分析 ............................................................................................... 24
3.4 本章小结 .......................................................................................................... 27
第四章 年金组合定价及风险评估 .............................................................................. 29
4.1 年金组合定价 .................................................................................................. 29
4.2 随机模拟方法逼近分布函数及分位数 .......................................................... 30
4.2.1 随机模拟方法 ....................................................................................... 30
4.2.2 案例分析 ............................................................................................... 32
4.3 中心极限方法逼近分布函数及分位数 .......................................................... 34
4.3.1 中心极限法 ........................................................................................... 34
4.3.2 案例分析 ............................................................................................... 35
4.4 本章小结 .......................................................................................................... 37
第五章 剩余寿命模糊随机模型 .................................................................................. 38
5.1 传统剩余寿命测度方法 .................................................................................. 38
5.1.1 随机变量测度 ....................................................................................... 38
5.1.2 模糊语言变量和隶属函数 ................................................................... 40
5.2 剩余寿命的模糊随机变量方法 ...................................................................... 41
5.3 本章小结 .......................................................................................................... 43
第六章 总结和展望 ...................................................................................................... 45
6.1 研究内容总结 .................................................................................................. 45
6.2 展望 .................................................................................................................. 45
参考文献 ........................................................................................................................ 47
附 录 .............................................................................................................................. 51
在读期间公开发表的论文和承担科研项目及取得成果 ............................................ 55
致 谢 .............................................................................................................................. 56
第一章 绪 论
1
第一章 绪 论
1.1 选题背景及意义
随着经济的发展,人民生活水平逐渐提高,我国人口平均寿命逐渐增加。由于
历史及人口政策的原因,我国人口结构失衡,逐渐进入人口老龄化社会,产生“未
富先老”的状况。人口老龄化将会对社会和家庭造成影响,特别是对经济增长、
技术进步、储蓄和劳动生产率等经济因素都会产生负面影响。
当然,面对如此日益严重的问题,我国建立了社会保障制度,由国家、企业和
个人共同分担养老责任。所以,我国养老金制度主要由社会基本养老保险、企业
年金、商业养老保险这“三大支柱”组成。
根据国家统计局的人口统计数据显示,从 1990 年到 2013 年,老年人口比例(本
文统计数据为 65 岁以上人口占总人口比)从 5.57%上升的 9.7%。而劳动力人口的
比例并未同比例增加,这样会带来一个重大问题,传统的社会基本养老保险将会
产生巨大偿付压力。
当“第一支柱”不足以满足养老的需求时,必须发展职业年金(包括私人退休
金和企业年金)和商业养老保险。历史上,每种新技术的来临,总是会掀起一阵
革命狂潮。瓦特的蒸汽机引领世界进入工业时代,电气技术的掌握使人类开始了
第二次工业革命,计算机的发明以及网络的普及让我们进入了互联网新时代。新
时代的来临总会对传统行业和传统技术带来冲击,人们总是要改进自身或者组织
的机制,以适应新环境。旧的体制总是要不断改革,以满足生产力发展的需求。
由于我国退休养老金双轨制的存在,关于养老金制度的改革是近年来人们热烈
议论的话题。究其原因,制度因素除外,一个重要因素是年金精算现值估计不准
确。而对于一般寿险产品,准确定价其毛费率将保证保险公司对保险市场获得有
利的竞争地位。
生命年金合同的定价模型涉及到两种不确定因素:年金服务对象的死亡率和随
着金融市场波动的利率。不确定性主要变现为随机性和模糊性,对于死亡率,一
般的研究是基于人口统计学上的生存模型进行研究,从而构建随机死亡模型。对
于利率,往往存在这样的模糊性术语,例如,未来
n
年的利率预期为“比较高”或
“比较低”,利率的不确定性表现为模糊性。由于死亡率的随机性和利率的模糊性,
年金现值精算过程是不确定的过程。因此,我们需要更可靠的理论来支持年金精
算现值理论。
相关推荐
作者:侯斌
分类:高等教育资料
价格:15积分
属性:59 页
大小:3.38MB
格式:PDF
时间:2025-01-09