An Empirical Study On The Coefficient Of Time-varying Beta Based On Difference Frequency Of Samples

Capital Asset Pricing Model (CAPM) has been widely used in the valuation of securities and portfolios. The most significant factor is beta coefficient, which describes the influence of individual assets affected by the overall market volatility. And beta is a measure of systemic risk securities. Since the parameter beta plays an important role in investment theory and practice, so the accurate estimation of beta is very important in theoretical and practical.By analyzing the literature about estimate the parameter beta, there are two major shortcomings:Firstly, the mainstream calculation method is still based on low frequency data, the use of high-frequency data is relatively small. The impact of information on the asset prices is a continuous process, when the data sampling frequency is relatively low, the low frequency data lost a lot of useful information. High-frequency data contains more useful information than the low-frequency data, and contains information will increase with the increase of sampling frequency. Secondly, with the increase of the sampling frequency, high-frequency data contained more and more rich information, scholars have put a lot of high-frequency covariance matrix estimation methods. Some of these method has been applied to the estimation of the parameters beta. But the impact of market microstructure noises will be more obvious. Moreover, return on assets will undergo substantial fluctuations in short span of time under certain circumstances, resulting in jumping; jumps will also cause no ignorable influence on the estimation of high-frequency covariance matrix. However, most of the existing literatures have only studied how to reduce the impact of the market microstructure noises on the estimation of high frequency covariance matrix, or only considered how to eliminate jumps effects on high frequency data. There is little literature taking into account the impact of the noises and jumps on the estimation of high frequency covariance matrix simultaneously.This paper use a new high-frequency covariance matrix estimator to eatimate beta coefficient—Modified Threshold Pre-averaging Realized Covariance Matrix (MTPCOV), this estimator can reduce the impact of microstructure noises by pre-averaging method, and can exclude the impact of jumps.To illustrate this method is more accurate in estimating beta. This paper also choose:Kalman filtering algorithm based on state space model, DCC-MVGARCH model and KCOV estimation method to estimate the parameters of beta. Establish a beta constraints by using the results of the four methods and build dynamic portfolio of assets.To analyze the effectiveness of Time-varying beta constraints by comparing the effect of dispersion portfolio unsystematic risk and maximize portfolio returns. We can know which method is more accurate when estimating the parameters beta. The paper results show that MTPCOV is better than other method.