Font Size: a A A

Dynamic Asset Allocation Based On Conversion Of Stock Status

Posted on:2012-05-07Degree:MasterType:Thesis
Country:ChinaCandidate:X WuFull Text:PDF
Abstract/Summary:PDF Full Text Request
It is set by the conventional average return-variation and CAPM that the optimal portfolio consists of constant weights of various assets across different period regardless any change of external market environment, which is contradictory to reality apparently. The inner average return and deviation of the stock market change with fluctuations of economic and other factors, which would impact the optimal portfolio choice. As a result, it is of vital importance to incorporate the changing external environment as an indispensable factor into asset allocation.The meaning of this paper lies in two aspects:on one hand, by using Markov Switching model, this paper take into account economic environment changes in the CAPM estimation, in order to improve the CAPM model in practice of applicability. On the other hand, allocating different assets in term of various dynamic states offers theoretical basis for dynamic asset allocation decisions.The contribution of this paper lies that it is the first time to combine the MR-VAR model and asset allocation in A-share market on the basis of latest market data, filling the gap in the absence of China's scholars in the relevant research fields, since previous researches on MR-VAR model only focus on market environment specification while ignoring asset allocation performance using this method.This paper divides stock market return dynamics into three states, namely, bear state, slow bull state and maniac bull state by using a Markov regime switching model. Then we build a conditional CAPM model, in which the stock market state being looked as a conditional variable. Based on the result of the conditional CAPM, we use mean-variation optimizing method to get the optimal portfolio under the three regimes. At last, we evaluate the performance of the dynamic asset allocation strategy. The empirical results show that because of making full use of the information reflecting time-varying economic environment. The dynamic asset allocation strategy based on Markov regime switching of stock market states is superior to the single-state asset allocation strategies.
Keywords/Search Tags:time-varying, dynamic asset allocation, Markov regime switching, performance evaluation
PDF Full Text Request
Related items