Academic Journal of Computing & Information Science, 2020, 3(3); doi: 10.25236/AJCIS.2020.030310.
School of Science, Hangzhou Normal University, Zhejiang Hangzhou, 311121
This paper obtained all the data of Kweichow Moutai stock from its listing date to July 2,2020 from Yahoo Finance, took the logarithm of the daily closing price and then differentiated to obtain the daily logarithm return rate to approximately replace the daily return rate, and modeled the time series of the daily logarithm return rate of Kweichow Moutai stock.The ARMA (0,(2)) sparse coefficient model of the daily log return rate is established by combining the ACF and BIC graphs of the daily log return rate. The residual square test of the model results shows that there is a strong autocorrelation. Therefore,based on this, the ARMA(0,2)+GARCH(1,1) model is improved according to the experience.By residuals and residuals squared tests of the model results, it is found that all of them are stationary white noise sequences. Therefore, it is believed that the improved model has a high degree of extraction of the original sequence information and a good fitting effect. Moreover, this model is used to predict the daily logarithm return rate in the next 6 days.
time series analysis, ARMA model, GARCH model, Kweichow Moutai Stock, Daily log rate of return
Di Jin. Kweichow Moutai Stock Daily Return Forecast Analysis. Academic Journal of Computing & Information Science (2020), Vol. 3, Issue 3: 92-100. https://doi.org/10.25236/AJCIS.2020.030310.
 https://www.jianshu.com/p/653d0f01531d?Clicktime = 1578736149
 R Language and Modern Statistical Methods, edited by Liu Qiang et al., Tsinghua University Press