时变风险度量模型

The metric method of time-varying risk

自从Markowitz在1952年发表"Portfolio Selection"以来,金融学家已设计了不少的风险度量方法,如方差法、下半方差法、平均偏差法、最大偏差法、VaR,等等.然而,这些方法有一个共同点,那就是,只考虑证券的价格运动,而没有将交易量因素考虑到风险度量中去.在通常情况下,这些风险度量方法以及建立在这些风险度量上的金融理论模型是能帮助投资者有效地防范和分散风险,但对类似于1997年的东南亚金融危机和2007年由美国次贷危机引发的金融海啸等一些突发事件所引起的金融危机却不能进行有效地防范,基于此,本文提出了一个新的风险度量指标——时变风险,以求既能反映通常情况下的风险状况,又能对突发事件的影响有所反映,以达到有效防范各种情况下的风险.时变风险是一个时间的函数,它是随时间变化的量,不是一个常量,这与人们对风险的感性认识是一致的,因此它对实践具有重大的指导意义.

Since Harry. M. Markowitz published ＂Portfolio Selection＂ 1952, financial academicians have proposed many methods to measure risk such as variant, downside-variant, average absolute deviation, maximum deviation, VaR and so on. However, these methods have a common limitation. It is that they only consider the price moment. Because they don＇t consider the influence of exchange quantity, they can＇t effectively keep away the risk caused by unexpected event such as Southeast Asia financial crisis 1997 and recent global financial tsunami. In this paper, a new method of measuring risk which called time-varying risk has proposed. The goal of this method is to respond the usual risk as well as the risk caused by unexpected event. The time-varying risk is a kind of variable with time, not a constant, which is identical to people＇s sense to risk, so it has significant meaning to practice.