中小板ETF与创业板ETF的统计套利探析
|
摘要
统计套利是一种在对过往的信息进行分析的基础之上,通过对变量概率分布的估计,并结合对基本面信息的分析,从而建立一种能够获得稳定获利的投资模型,实现降低资产价值的波动性并且提高收益率的目的。本文在明确统计套利的本质和理论依据的基础上,选定创业板ETF和中小板ETF进行两者之间的统计套利,通过对协整关系的检验确定两个标的之间的关联性,进行了基于两者价差均值回归的实证分析,并且根据历史数据交易价格的规律,提出最优化套利模型,进而探讨在目前的市场环境下能否通过在创业板ETF和中小板ETF之间进行配对交易实现低风险套利。
Statistical arbitrage is based on the analysis of the past information,through the estimation of the probability distribution of variables,and combined with the analysis of fundamental information,so as to establish a stable profitable investment model,realize to reduce the volatility of asset value and improve the purpose of the rate of return.Based on the nature and theoretical basis of statistical arbitrage,the paper chooses the gem ETF and SME board ETF to arbitrage the statistical arbitrage between the two indicators,determines the correlation between the two objects,makes empirical analysis based on the average regression of the two prices,and according to the law of historical data transaction price,puts forward the optimal arbitrage model,and then discusses whether can realize the low risk arbitrage through the gem ETF and SME board ETF in the current market environment.
Statistical arbitrage is based on the analysis of the past information,through the estimation of the probability distribution of variables,and combined with the analysis of fundamental information,so as to establish a stable profitable investment model,realize to reduce the volatility of asset value and improve the purpose of the rate of return.Based on the nature and theoretical basis of statistical arbitrage,the paper chooses the gem ETF and SME board ETF to arbitrage the statistical arbitrage between the two indicators,determines the correlation between the two objects,makes empirical analysis based on the average regression of the two prices,and according to the law of historical data transaction price,puts forward the optimal arbitrage model,and then discusses whether can realize the low risk arbitrage through the gem ETF and SME board ETF in the current market environment.
引文
[1]DaiJin.Statistical arbitrage of stock index futures and ETF based on cointegration[J].Securities&Futures of China,2012(10).
[2]杨文涛,罗天勇,贺颖,李伟,何冰.基于ETF统计套利的实证研究[D].贵州财经大学,2013.
[3]蔡燕,王林,许莉莉.基于随机价差法的配对交易研究[J].金融改革,2012(08).
[4]李育补.ETF配对套利机会影响因素研究[D].哈尔滨工业大学,2015.
[5]刘华.基于统计套利的ETF期现套利方法应用研究[D].大连理工大学,2015.
[6]谢洪军,朱晓莉,张慧.我国发电量与工业增长协整关系实证分析[J].科技和产业,2017(01).
(1)令Mspreadt为价差序列,σ为价差序列标准差,k为待求参数。设套利收益的期望函数为:E(kσ)=λ·R(kσ)·ψ-1(kσ)。其中:λ为一常数,R(kσ)=kσ为阈值取kσ时的可套利空间:λ(x)={kψ|p(|Mspreadt)|≥kσ=x}。所以ψ-1(kσ)表示|Mspreadt|>σ的概率。k的最优取值为k={y|E(kσ)≤E(yσ),k∈(0,2)}。
[2]杨文涛,罗天勇,贺颖,李伟,何冰.基于ETF统计套利的实证研究[D].贵州财经大学,2013.
[3]蔡燕,王林,许莉莉.基于随机价差法的配对交易研究[J].金融改革,2012(08).
[4]李育补.ETF配对套利机会影响因素研究[D].哈尔滨工业大学,2015.
[5]刘华.基于统计套利的ETF期现套利方法应用研究[D].大连理工大学,2015.
[6]谢洪军,朱晓莉,张慧.我国发电量与工业增长协整关系实证分析[J].科技和产业,2017(01).
(1)令Mspreadt为价差序列,σ为价差序列标准差,k为待求参数。设套利收益的期望函数为:E(kσ)=λ·R(kσ)·ψ-1(kσ)。其中:λ为一常数,R(kσ)=kσ为阈值取kσ时的可套利空间:λ(x)={kψ|p(|Mspreadt)|≥kσ=x}。所以ψ-1(kσ)表示|Mspreadt|>σ的概率。k的最优取值为k={y|E(kσ)≤E(yσ),k∈(0,2)}。