Research Of Two Problems In Financial Markets

The thesis is divided into two parts,The first part is a research on the relation between the stock return and the investment rate.we randomly selected30stocks from the Chinese securities market. We carried out research about the relation between the stock yield rate and the investment rate in the laws of the statistics, using its investment rate reports and stock prices from December25th in2012to December25th in2013. Among them, in order to ensure the validity of the statistical analysis, we use the famous Box-Cox Transformation in the statistical analysis and adopt the optimization method. By maximizing the F-statistic, we could choose the optimal linear regression model among the stock yield rate, the invest-ment rate and the previous stock yield. The empirical analysis results show that the correlation between stock returns and its early returns, however, the analysts’investment rate has nothing to do with the subsequent performance of the stock yield rate.The second part is a research of a new GARCH model parameter estimation methods. Quantile GARCH is a kind of new time series models,aiming to some statistical features of the financial markets.As soon as putting forward,it gain the recognition and praise of the research community,but we have not yet found a good way to work out the parameter estimation problem,this will affect the application of the model,such as option pricing and the optimal portfolio selection.This research mainly based on the new GARCH model parameters in two parts,namely,the regression model of variance sequence and the marginal distribution.In the process of tackling the pa-rameter estimation,we use a new method which proposed by Qiwei Yao who put forward to estimate the general GARCH model.Firstly,we will conduct the appropriate standardization with the model,using the standardized sequence to estimate the coefficient of variance regression model.Secondly,based on the esti-mated coefficient,we will start secondary standardized sequence,immediately,the standardized sequence is used to estimate the parameters in the marginal distribution.In the course of researching,we get in-depth analysis of the parameters of the model structure and absorb the merits of the method proposed by Qiwei Yao.We succeeded in dividing the parameter estimation problem of the original model into two problems of low dimension,what’s more, good estimation results have been achieved.Also,we carry out the numerical simulation test by applying Matlab in this paper.