Multi-scaling Research On β Coefficients In Chinese Stock Market

Modern financial risk management and many other classical financial theories and models are based on Fama's Efficient Market Hypothesis. However, many anomalies have been found in worldwide financial market in the later 1970s. Recently, many foreign and domestic empirical studies have demonstrated that the fluctuation of financial time series show not only chaos and fractal features but also obvious multifractal characteristics ,that is, the price fluctuation characteristics at different time scales is various. According to the hypothesis that financial market is multifractal, this paper researches the multi-scaling characteristics of theβcoefficients.Firstly, this paper reviews the stock price fluctuation theories, and analyses the multifractal characteristics that probably exist in china financial market. Secondly, according to the hypothesis, proposes a multi-scaling estimation model based on wavelet theories, and make significance test ofβ's multi-scaling characteristics under different market conditions, the conclusions support the multifractal hypothesis. Furthermore, the paper researches how the trading frequency and risk style factors influenceβ's multi-scaling characteristics.The empirical study finds that theβ's multi-scaling characteristics is significant at bear market, the more frequently of trading, the more significant ofβ's multi-scaling characteristics but it has no apparent trends, trading frequently magnifies the function of signal transfer to the extent, it makes stock price overreact to bad news. Generally speaking, along with the scale rising, high risk stock'sβis increase, low risk stock'sβis decrease, but in bear market, low risk stock'sβis fluctuates in line with scale rising, stock price is more sensitive to bad news than good news. All of these reflect the investors in china are immature and blind, and bring extremely strong speculation. It is also proved that Chinese security market is a non-effective market.