本文關(guān)鍵詞：基于動(dòng)態(tài)收縮法的大維協(xié)方差陣的估計(jì)及其應(yīng)用　出處：《統(tǒng)計(jì)與決策》2017年10期 　論文類型：期刊論文

  更多相關(guān)文章： 大維協(xié)方差陣 動(dòng)態(tài)加權(quán)收縮估計(jì)量 投資組合

【摘要】：文章將單因子協(xié)方差陣和樣本協(xié)方差陣相結(jié)合,通過對(duì)它們進(jìn)行最優(yōu)加權(quán)平均,提出了新的協(xié)方差陣估計(jì)方法——?jiǎng)討B(tài)加權(quán)收縮估計(jì)量(DWS)。該估計(jì)量一方面通過選擇最優(yōu)的權(quán)重來平衡協(xié)方差陣估計(jì)的偏差和誤差;另一方面估計(jì)的是大維數(shù)據(jù)的動(dòng)態(tài)協(xié)方差陣,在估計(jì)過程中考慮了前期信息的影響。通過模擬和實(shí)證研究發(fā)現(xiàn):較傳統(tǒng)的協(xié)方差陣估計(jì)方法而言,DWS估計(jì)量明顯提高了大維協(xié)方差陣的估計(jì)效率;并且將其應(yīng)用在投資組合時(shí),投資者獲得了更高的收益和經(jīng)濟(jì)福利。
[Abstract]:In this paper, the univariate covariance matrix and the sample covariance matrix are combined to carry out the optimal weighted average. A new covariance matrix estimation method, dynamic weighted contraction estimator, is proposed. On the one hand, the deviation and error of covariance matrix estimation are balanced by selecting the optimal weight. On the other hand, the dynamic covariance matrix of large dimensional data is estimated, and the influence of pre-information is taken into account in the estimation process. Through simulation and empirical research, it is found that: compared with the traditional covariance matrix estimation method. The DWS estimator obviously improves the estimation efficiency of the large dimensional covariance matrix. And when applied to the portfolio, investors get higher returns and higher economic benefits.
【作者單位】： 貴州財(cái)經(jīng)大學(xué)數(shù)學(xué)與統(tǒng)計(jì)學(xué)院;德雷賽爾大學(xué)LeBoro商學(xué)院;
【基金】：國家社會(huì)科學(xué)基金青年項(xiàng)目(16CTJ013) 全國統(tǒng)計(jì)科學(xué)研究項(xiàng)目(2015LY19) 貴州省教育廳普通本科高校自然科學(xué)研究項(xiàng)目(黔教合KY字[2015]423)
【分類號(hào)】：F224;F830.9
【正文快照】： 0引言近年來,圍繞如何估計(jì)大維協(xié)方差陣的估計(jì)問題,已經(jīng)引起了學(xué)者們的廣泛關(guān)注。目前的研究主要分為三類,第一類是稀疏協(xié)方差陣估計(jì)方法(Bickel和Levina(2008)[1],Rothman(2012)[2],Lam和Fan(2009)[3],Cai和Zhou(2012)[4],Cai和Liu(2011)[5]等),該方法假定有些資產(chǎn)之間是不相，

本文編號(hào)：1412018