【摘要】 本文创造性地指出,股价波动具备先增长后下降的倒U形曲线特征,即库兹涅茨曲线特征,从而对最高值出现的时间点进行预测.本文首先对股价波动曲线进行了去除微小波动和类均线的光滑化处理,通过多项式函数回归和Logistic函数回归的拟合方法,来确定曲线增长部分的拐点,并给出寻找拐点和预测的方法.抽样检测结果表明该方法非常具有实效性.更多还原

【Abstract】 This paper shows that stocks price fluctuations have the characteristics of Kuznets Curve,which can predict the time that a stock price reaches its maximum during a period.The polynomial function fitting and logistic function fitting are used for estimation,which is for prediction of the inflection point.The real data analysis shows that the proposed method performs well.更多还原