Analysis On Multifractral Cross-correlation Features Of A+H Cross-listing Stocks

With the opening of "Shanghai-Hong Kong Stock Connect" and "Shenzhen-Hong Kong stock connect",the stock markets between mainland and Hong Kong are closely linked and the form of A+H cross-listing is more respected by the financial field.In addition,as the frontier of finance physics,the multifractal detrending cross-correlation method is more and more widely used in the financial field.In this paper,79 sample stocks cross-listed by A+H are selected from January 1,2004 to May 26,2017,and A-share and H-share market indices are compiled correspondingly.Based on the simulation analysis of the multi-fractal cross-correlation method,the overall and asymmetric multi-fractal cross-correlation method are utilized to analyze the cross-correlation of the A-share market and the H-share market and the direction of risk transmission.The multifractal detrended cross-correlation method under different filtering is used to measure the long-term memory feature and the accuracy of the multifractal feature of the empirical object.Simulation analysis find that in both cases of Hxy<1/2(Hx+Hy)and Hxy=1/2(Hx+Hy),multifractal detrending cross-correlation analysis based on maximum overlap wavelet transform(MFDCCA-MODWT)is optimal for measuring long-term memory features,while multifractal detrend cross-correlation moving algorithm(MFXDMA-0 and MFXDMA-1)perform best when measuring multifractal features,and MFDCCA-MODWT performs the worst.Then these two methods were used to measure the long memory characteristics and multifractal features of China's A+H cross-listed stocks respectively on the whole.It was found that there is a long memory between A+-H shares in China,and there is a weak multifractal features cross-correlation between them.The multifractal feature is mainly caused by the long-range correlation of the sequence,and has nothing to do with the fat-tail distribution of the sequence.Besides,in this paper,the asymmetric MF-DCCA and the MF-DCCA method based on time lag are used to analyze the asymmetric cross-correlation characteristics and risk transmission of the A+H shares from the different rising-downward trend and the different fluctuation direction.The empirical results show that the A+H shares both have long-memory in different rising and decreasing trend,and the long memory characteristic is stronger in the downward trend of stock price.In addition,we also found that there is a two-way risk conduction effect between A + H shares and the mainland market is more conductive to the Hong Kong market.