本文選題：金融資產(chǎn) + 證券市場　； 參考：《西南交通大學(xué)學(xué)報(社會科學(xué)版)》2017年04期

【摘要】：利用日內(nèi)高頻交易數(shù)據(jù)對金融資產(chǎn)收益率的波動率進行建模分析是近年來理論界和實務(wù)界共同關(guān)注的熱點問題。以2005年初到2016年底我國日間HS300指數(shù)5分鐘高頻數(shù)據(jù)為樣本,實證分析以漸進理論為基礎(chǔ)的ARQ及HARQ模型對我國股票市場高頻波動的預(yù)測效果,并與跳躍、跳躍變差及正負向跳躍變差為基礎(chǔ)的HAR-RV、HAR-JC、CHAR及SHAR等多種高頻波動率預(yù)測模型進行比較,發(fā)現(xiàn):ARQ及HARQ模型對我國股票市場具有更高的預(yù)測精度。
[Abstract]:The modeling and analysis of volatility of financial asset return by using intraday high-frequency trading data is a hot issue in theory and practice in recent years. Based on the 5-minute high-frequency data of China's daytime HS300 index from the beginning of 2005 to the end of 2016, this paper empirically analyzes the prediction effect of the ARQ and HARQ models based on the progressive theory on the high-frequency volatility of China's stock market, and jumps. Based on the jump variation and the positive and negative jump variation, several kinds of high-frequency volatility prediction models, such as HAR-RVN HAR-JCNCHAR and SHAR, are compared. It is found that the two models have higher prediction accuracy for China's stock market.
【作者單位】： 西南交通大學(xué)經(jīng)濟管理學(xué)院;
【分類號】：F224;F832.51
，

本文編號：1911179