基于长短记忆神经网络预测的大类资产因子配置研究

大类资产配置策略为资金相对庞大的机构投资者提供了一个有效获取稳健收益的手段,通过因子收益分布及相关性的预测能更好地进行大类资产配置.将大类资产因子配置的思想与机器学习算法预测有机结合,首先筛选宏观因子及风格因子,利用长短记忆神经网络(LSTM)方法预测组合收益,得到最优因子组合;然后结合最优因子组合中蕴含的信息,修正对资产预期收益率的估计并提出了大类资产的权重配置方案.通过在全球18种大类资产上进行的算例分析表明,采用本文模型得到的配置策略相较于其它模型有更高的收益风险比、较低的年化波动以及较小的最大回撤.研究结论可为机构投资者的大类资产配置提供理论借鉴.     

考虑高频数据影响的BEKK模型的估计及其应用

多维资产的协方差阵在投资组合中扮演着重要角色,如何估计和预测资产的协方差阵是统计领域的一大热点问题.将基于高频数据的已实现协方差阵(RCOV)和双频已实现协方差阵(TSCOV)应用到BEKK模型的估计过程中,提出了考虑高频数据影响的BEKK-RCOV和BEKK-TSCOV模型,这两类模型将高频数据引入到协方差阵估计过程中的同时,还可以对协方差阵直接进行预测,避免了预测模型的选择困难问题,并且提高了协方差阵的估计效率.通过实证研究发现:BEKK-RCOV和BEKK-TSCOV模型估计和预测效果明显优于BEKK模型,将其应用在投资组合时,使投资者获得了更高的收益.     

上证综指的已实现波动率预测模型

采用上证综指2000-2008年的高频数据,在考察了中国股市已实现波动率的特征(即具有长记忆性、结构突变、不对称性和周内效应的特征并且结构突变只能部分解释已实现波动率的长记忆性)的基础上,构建了一个自适应的不对称性HAR-D-FIGARCH模型,并用于波动率的预测。模型的估计结果表明,与其他HAR模型相比,该模型对样本内数据的拟合效果最好。最后,通过SPA检验实证评价和比较了该模型与其他5种已实现波动率预测模型的样本外预测精度。结果发现,在各种损失函数下,该模型是预测中国股市已实现波动率精度最高的模型。     

宏观经济对股市波动的影响——基于GARCH-MIDAS-RTSRV模型的证据

本文对GARCH-MIDAS模型进行了拓展。首先,在估计GARCH-MIDAS模型的长期波动成分时,采用同时考虑噪声和跳跃影响的稳健双频已实现波动估计量RTSRV来代替传统的已实现波动估计量RV。其次,选取了经济变量并从中提取出主成分,从其水平值和波动率两个层面研究不同主成分对股市波动的影响。研究发现：本文构造的GARCH-MIDAS-RTSRV模型优于传统的GARCH-MIDAS模型,其预测精度更高并且可使投资者获得更高的经济价值;经济变量的主成分和已实现波动率均对股市的波动有显著的影响,并且相较于其水平值,波动率对股市波动的影响更为显著。     

基于数据分解集成和高频数据建模的汇率波动率预测

汇率波动率是刻画外汇金融资产收益变化程度的指标，也是度量外汇风险的方法之一，汇率波动对经济与金融系统都有重要的影响。由于非平稳和非线性的特征，准确预测汇率波动率一直是金融研究的重点和难点。为了提高预测汇率波动率的准确性，本文采用基于人民币汇率高频数据计算的已实现波动率和机器学习方法，对数据进行分解集成和建模，提出了一种有效的多尺度EEMD-PSR-SVR-ARIMA预测模型。具体过程如下：首先，采用集合经验模态分解(EEMD)的方法将复杂的时间序列分解成不同尺度的本征模态函数和趋势项；然后采用支持向量回归(SVR)的方法对本征模态函数进行预测，并利用相空间重构和粒子群优化的方法来确定SVR模型的输入维数与参数。同时，使用差分自回归移动平均模型(ARIMA)预测趋势项；最后集成得到模型预测的结果。实证结果表明EEMD-PSR-SVR-ARIMA模型可以有效地提高汇率波动率预测的精度。     

基于非对称双成份CARR模型的波动率预测

研究表明,基于日内(高低价)数据构建的价格极差测度相比日度收益率包含更多关于真实波动率的信息,同时,波动率具有聚集性、非对称性和长记忆性等丰富、复杂的典型特征,综合考虑这些特征对波动率进行建模与预测非常重要。本文在对价格极差建模的CARR模型的基础上,对其进行扩展,构建了双成份CARR (CCARR)模型来对波动率进行预测。该模型假设价格极差的条件均值由两个成份组成,即长期成份与短期成份.该模型能够捕获波动率长记忆性,且容易进一步扩展为非对称CCARR (ACCARR)模型来捕获杠杆效应(波动率非对称性)。(A)CCARR模型具有较高的建模灵活性,且易于实现。采用上证综合指数、香港恒生指数、日本Nikkei225指数、法国CAC40指数和德国DAX指数数据进行实证分析,以价格极差与已实现波动率(RV)作为比较基准,四种预测评价指标及Mincer-Zarnowitz检验结果表明:杠杆效应与双成份极差(波动率)都对样本外波动率预测具有重要影响,且杠杆效应相比双成份极差对于样本外波动率预测的影响更大;考虑了杠杆效应的双成份ACCARR模型具有最好的样本外波动率预测效果,其次是ACARR模型,CARR模型表现最差。     

部分线性模型在贵州省公路货运量预测中的应用

运用部分线性模型对贵州省公路货运量进行预测研究.首先运用灰色关联度分析法确定影响贵州省公路货运量的主要影响因子;然后运用主成分分析法将选取的影响因子指标数据进行降维处理,通过分析处理后的数据得到部分线性模型;最后,以2010-2012年的公路货运量作为验证值,将部分线性模型、多元线性回归模型及灰色预测模型的预测结果进行比较.研究结果表明:部分线性模型能较好地拟合贵州省1990-2009年公路货运量;三种模型的预测结果显示,部分线性模型预测结果优于多元线性回归模型和灰色预测模型的预测结果.     

基于高频波动率模型与R-vine copula的行业资产组合风险测度研究

相较于低频波动率模型,高频波动率模型在单资产的波动和风险预测中均取得了更好效果,因此如何将高频波动率模型引入组合风险分析具有重要的理论和现实意义。本文以沪深300指数中的6种行业高频数据为例,运用滚动时间窗技术建立9类已实现波动率异质自回归(HAR-RV-type)模型刻画行业指数波动,同时使用R-vine copula模型描述行业资产间相依结构,进一步结合均值-CVaR模型优化行业资产组合投资比例,构建组合风险的预期损失模型,并通过返回测试比较不同风险模型的精度差异。研究结果表明:将HAR族高频波动率模型引入组合风险分析框架,能够有效预测行业资产组合风险状况;高频波动率预测的准确性将进而影响组合风险测度效果,跳跃、符号跳跃变差以及符号正向、负向跳跃变差均有助于提高行业组合风险的预测精度。     

考虑微观结构噪声与跳跃影响的波动建模

现有的金融高频数据研究,并未充分考虑微观结构噪声对波动建模和预测的影响.以非参数化方法为理论框架,基于高频数据,采用适当方法分离出波动中的微观结构噪声成份,构建了新的跳跃方差和连续样本路径方差,将已实现波动分解为连续样本路径方差、跳跃方差和微观结构噪声方差.同时考虑微观结构噪声和跳跃对波动的影响,对HAR-RV-CJ模型进行改进,提出了HAR-RV-N-CJ模型和LHAR-RV-N-CJ模型.通过上证综指高频数据进行实证,结果表明新模型在模型拟合和预测方面均优于HAR-RV-CJ模型.     

基于ARIMA与信息粒化SVR组合的股指预测研究

对于证券市场投资者而言,基于合理假设准确预测资产价格未来发展方向与趋势关乎投资成败。本文通过构建一个基于ARIMA与信息粒化SVR的组合预测模型,对股票市场指数价格和收益变化的趋势进行预测。实证研究结果表明:基于ARIMA与信息粒化SVR组合的股指预测模型相较于传统时间序列模型而言,在预测精度和效度方面有较大提升,能够在一定时间周期内对股票等风险资产的价格波动区间进行较为可靠地预测,但目前还只能大致确定时间序列波动的区间范围而不能精确地预测具体点位。未来仍需结合其他预测模型和预判技术进一步深入研究,以有效提升股指趋势预测的准确性和实际指导性。     

Volatility forecasting for interbank offered rate using grey extreme learning machine: The case of China

Interbank Offered rate is the only direct market rate in China’s currency market. Volatility forecasting of China Interbank Offered Rate (IBOR) has a very important theoretical and practical significance for financial asset pricing and financial risk measure or management. However, IBOR is a dynamics and non-steady time series whose developmental changes have stronger random fluctuation, so it is difficult to forecast the volatility of IBOR. This paper offers a hybrid algorithm using grey model and extreme learning machine (ELM) to forecast volatility of IBOR. The proposed algorithm is composed of three phases. In the first, grey model is used to deal with the original IBOR time series by accumulated generating operation (AGO) and weaken the stochastic volatility in original series. And then, a forecasting model is founded by using ELM to analyze the new IBOR series. Lastly, the predictive value of the original IBOR series can be obtained by inverse accumulated generating operation (IAGO). The new model is applied to forecasting Interbank Offered Rate of China. Compared with the forecasting results of BP and classical ELM, the new model is more efficient to forecasting short- and middle-term volatility of IBOR.     

Medium‐term horizon volatility forecasting: A comparative study

In this paper, volatility is estimated and then forecast using unobserved components‐realized volatility (UC‐RV) models as well as constant volatility and GARCH models. With the objective of forecasting medium‐term horizon volatility, various prediction methods are employed: multi‐period prediction, variable sampling intervals and scaling. The optimality of these methods is compared in terms of their forecasting performance. To this end, several UC‐RV models are presented and then calibrated using the Kalman filter. Validation is based on the standard errors on the parameter estimates and a comparison with other models employed in the literature such as constant volatility and GARCH models. Although we have volatility forecasting for the computation of Value‐at‐Risk in mind the methodology presented has wider applications. This investigation into practical volatility forecasting complements the substantial body of work on realized volatility‐based modelling in business. Copyright © 2007 John Wiley & Sons, Ltd.     

基于价格极差的GARCH模型

本文利用资产价格的极差序列,基于常规GARCH模型的框架,构造了一类关于波动率的新模型,即GARCH-R模型以及能够表达波动率变化非对称性特性的AGARCH-R模型。利用上证综合指数日收益率及相应的高频数据,通过比较不同模型对波动率以及VAR的预测效果,揭示了这种包含了极差信息的新的模型比传统的GARCH类模型的预测效果具有显著的优势。     

多维金融高频协方差阵预测模型的比较分析

现代投资组合理论大部分是从组合风险控制的角度展开,协方差矩阵扮演着非常重要的角色.将高频协方差阵应用在投资组合或风险管理时,就需要考虑采用何种预测模型来对高频协方差阵进行预测,较好的预测模型能够更加准确的对资产的波动性进行预测.高频协方差阵预测模型的建立较为复杂,目前还没有一种广泛被认可的模型.采用MCS检验法来选择最优的预测模型,研究发现高频协方差阵预测模型LOG-HAR模型在所有的损失函数下预测能力最好,并且高频协方差阵预测模型的预测能力要优于低频协方差阵预测模型.     

Optimal portfolio choice and stochastic volatility

In this paper we examine the effect of stochastic volatility on optimal portfolio choice in both partial and general equilibrium settings. In a partial equilibrium setting we derive an analog of the classic Samuelson–Merton optimal portfolio result and define volatility‐adjusted risk aversion as the effective risk aversion of an individual investing in an asset with stochastic volatility. We extend prior research which shows that effective risk aversion is greater with stochastic volatility than without for investors without wealth effects by providing further comparative static results on changes in effective risk aversion due to changes in the distribution of volatility. We demonstrate that effective risk aversion is increasing in the constant absolute risk aversion and the variance of the volatility distribution for investors without wealth effects. We further show that for these investors a first‐order stochastic dominant shift in the volatility distribution does not necessarily increase effective risk aversion, whereas a second‐order stochastic dominant shift in the volatility does increase effective risk aversion. Finally, we examine the effect of stochastic volatility on equilibrium asset prices. We derive an explicit capital asset pricing relationship that illustrates how stochastic volatility alters equilibrium asset prices in a setting with multiple risky assets, where returns have a market factor and asset‐specific random components and multiple investor types. Copyright © 2011 John Wiley & Sons, Ltd.     

A delayed stochastic volatility correction to the constant elasticity of variance model

The Black-Scholes model does not account non-Markovian property and volatility smile or skew although asset price might depend on the past movement of the asset price and real market data can find a non-flat structure of the implied volatility surface. So, in this paper, we formulate an underlying asset model by adding a delayed structure to the constant elasticity of variance (CEV) model that is one of renowned alternative models resolving the geometric issue. However, it is still one factor volatility model which usually does not capture full dynamics of the volatility showing discrepancy between its predicted price and market price for certain range of options. Based on this observation we combine a stochastic volatility factor with the delayed CEV structure and develop a delayed hybrid model of stochastic and local volatilities. Using both a martingale approach and a singular perturbation method, we demonstrate the delayed CEV correction effects on the European vanilla option price under this hybrid volatility model as a direct extension of our previous work [12].     

Financial Risk Modelling and Econometric Inference

In finance, the explicit modelling of uncertainty takes on a particularly important role. The values of financial derivatives increase in the return volatility of the underlying security. This notion requires a concept of volatility and hence uncertainty. In addition, the choice between modelling in discrete and continuous time is not arbitrary, since it corresponds to a distinction between incomplete and complete markets, respectively, and this distinction matters for asset pricing, financial risk modelling, and inference. Risk and volatility are closely connected, and implied volatility, volatility forecasting, volatility in term structure models, stochastic volatility, and portfolio analysis are considered and related to a more general interplay between cross-sectional and dynamic aspects in finance. Stocks, bonds, and options are considered and placed in the context of efficiency and separation in inference.     

基于修正的已实现阈值幂变差的股市跳跃波动行为研究

基于非参数统计方法,利用考虑金融资产价格跳跃和杠杆效应的时点波动估计方法修正已实现阈值幂变差,构造甄别跳跃的检验统计量,对金融资产价格中的随机波动、有限活跃跳跃和无限活跃跳跃等问题进行综合研究。为同时吸收波动率的异方差集聚效应和收益率的非对称效应,对原有的已实现波动率异质自回归预测模型进行拓展,将非对称的异质性自回归模型的误差项设定为GARCH模型,以考察跳跃波动序列与连续波动序列之间的复杂关系。利用沪深股指高频数据进行实证研究,包括进行跳跃识别,跳跃活动程度检验和波动率预测效果对比。研究结果表明,沪深股市同时存在布朗运动成分、有限活跃跳跃和无限活跃跳跃成分,其中连续路径方差占主体。同时,收益和波动间的杠杆效应显著,无论短期还是长期,连续波动和跳跃波动对波动率的预测均具有显著影响,同时考虑股价的跳跃、波动和杠杆效应因素有助于更准确地刻画资产价格动态过程。     

基于Heston随机波动率模型和风险偏好视角的资产负债管理

本文研究基于Heston随机波动率模型的资产负债管理问题。假设金融市场由一个无风险资产和一个风险资产构成,投资者的目标是最大化其终端财富的期望效用。应用随机控制方法,得到了该问题最优资产配置策略的解析表达式和相应值函数的解析解,通过数值算例分析了Heston模型主要参数以及债务对最优资产配置策略的影响。结果表明:配置到风险资产的比例对Heston模型中的参数非常敏感;为了对冲债务风险,负债的引入使得配置到风险资产的比例比无负债情形下的高;在风险厌恶系数变大时,无论投资者是否有负债,其投资到风险资产的比例则越来越低。     

Common volatility and correlation clustering in asset returns

We present a new multivariate framework for the estimation and forecasting of the evolution of financial asset conditional correlations. Our approach assumes return innovations with time dependent covariances. A Cholesky decomposition of the asset covariance matrix, with elements written as sines and cosines of spherical coordinates allows for modelling conditional variances and correlations and guarantees its positive definiteness at each time t. As in Christodoulakis and Satchell [Christodoulakis, G.A., Satchell, S.E., 2002. Correlated ARCH (CorrARCH): Modelling the time-varying conditional correlation between financial asset returns. European Journal of Operational Research 139 (2), 350–369] correlation is generated by conditionally autoregressive processes, thus allowing for an autocorrelation structure for correlation. Our approach allows for explicit out-of-sample forecasting and is consistent with stylized facts as time-varying correlations and correlation clustering, co-movement between correlation coefficients, correlation and volatility as well as between volatility processes (co-volatility). The latter two are shown to depend on correlation and volatility persistence. Empirical evidence on a trivariate model using monthly data from Dow Jones Industrial, Nasdaq Composite and the 3-month US Treasury Bill yield supports our theoretical arguments.