基于UKF的非齊次泊松跳躍市場微結(jié)構(gòu)模型研究
發(fā)布時(shí)間:2018-08-31 09:06
【摘要】:本文針對不確定性因素引起資產(chǎn)價(jià)格的巨大波動(dòng),構(gòu)建了一個(gè)由非齊次泊松過程驅(qū)動(dòng)的跳躍市場微結(jié)構(gòu)模型.在模型參數(shù)未知的情況下,我們使用非參數(shù)化方法檢測出時(shí)變跳躍強(qiáng)度,由此再利用無跡卡爾曼濾波和極大似然法來估計(jì)跳躍市場微結(jié)構(gòu)模型參數(shù)的值.模擬仿真與實(shí)證分析驗(yàn)證了該方法的有效性,并利用AIC準(zhǔn)則對兩類跳躍波動(dòng)率模型進(jìn)行了優(yōu)劣比較.研究結(jié)果表明,跳躍市場微結(jié)構(gòu)模型在擬合股指數(shù)據(jù)方面要優(yōu)于跳躍隨機(jī)波動(dòng)模型.
[Abstract]:In this paper, a jumping market structure model driven by inhomogeneous Poisson process is constructed to deal with the huge volatility of asset price caused by uncertainty. When the model parameters are unknown, we use the nonparametric method to detect the time-varying jump intensity, and then use the unscented Kalman filter and the maximum likelihood method to estimate the parameters of the microstructural model of the jump market. The effectiveness of the proposed method is verified by simulation and empirical analysis, and two kinds of jump volatility models are compared with each other by AIC criterion. The results show that the jumping market microstructure model is superior to the jump stochastic volatility model in fitting stock index data.
【作者單位】: 長沙理工大學(xué)電氣與信息工程學(xué)院;國防科技大學(xué)信息系統(tǒng)與管理學(xué)院;中南大學(xué)信息科學(xué)與工程學(xué)院;長沙理工大學(xué)交通運(yùn)輸工程學(xué)院;
【基金】:國家自然科學(xué)基金(51507015) 中國博士后科學(xué)基金(2016M592949) 湖南省自然科學(xué)基金(2015JJ3008) 湖南省科技計(jì)劃(2015NK3035) 可再生能源電力技術(shù)湖南省重點(diǎn)實(shí)驗(yàn)室基金(2014ZNDL002)~~
【分類號】:F831.51;O211.6
本文編號:2214533
[Abstract]:In this paper, a jumping market structure model driven by inhomogeneous Poisson process is constructed to deal with the huge volatility of asset price caused by uncertainty. When the model parameters are unknown, we use the nonparametric method to detect the time-varying jump intensity, and then use the unscented Kalman filter and the maximum likelihood method to estimate the parameters of the microstructural model of the jump market. The effectiveness of the proposed method is verified by simulation and empirical analysis, and two kinds of jump volatility models are compared with each other by AIC criterion. The results show that the jumping market microstructure model is superior to the jump stochastic volatility model in fitting stock index data.
【作者單位】: 長沙理工大學(xué)電氣與信息工程學(xué)院;國防科技大學(xué)信息系統(tǒng)與管理學(xué)院;中南大學(xué)信息科學(xué)與工程學(xué)院;長沙理工大學(xué)交通運(yùn)輸工程學(xué)院;
【基金】:國家自然科學(xué)基金(51507015) 中國博士后科學(xué)基金(2016M592949) 湖南省自然科學(xué)基金(2015JJ3008) 湖南省科技計(jì)劃(2015NK3035) 可再生能源電力技術(shù)湖南省重點(diǎn)實(shí)驗(yàn)室基金(2014ZNDL002)~~
【分類號】:F831.51;O211.6
【相似文獻(xiàn)】
相關(guān)期刊論文 前1條
1 甘敏;彭輝;陳曉紅;;基于市場微結(jié)構(gòu)模型和進(jìn)化算法的資產(chǎn)分配[J];系統(tǒng)工程學(xué)報(bào);2011年03期
相關(guān)碩士學(xué)位論文 前1條
1 阮昌;基于狀態(tài)空間法的金融市場微結(jié)構(gòu)非線性動(dòng)力學(xué)建模及其數(shù)值模擬[D];中南大學(xué);2011年
,本文編號:2214533
本文鏈接:http://sikaile.net/kejilunwen/yysx/2214533.html
最近更新
教材專著