论文部分内容阅读
(二) 趋势外推法当时间序列的散点图呈比较规则的直线或曲线趋势变化时,可以用某些初等函数来描述时间序列的变化规律,并依时间的外推来进行预测,这种方法称为时间序列的趋势外推法。在市场预测中,通常选用的初等函数有一次函数(直线)、二次函数(抛物线)、指数函数、幂函数、修正指数函数以及龚珀资曲线、逻辑曲线等。由于篇幅有限,在此仅介绍市场预测中最常用的直线型和指数型趋势外推法。
(2) Extrapolation of Trends When the scattergram of a time series shows a trend of changes in a straight line or curve of a relatively regular rule, some elementary functions can be used to describe the variation of the time series, and extrapolation by time can be used for prediction. This method is called extrapolation of the trend of the time series. In market forecasting, the elementary functions that are usually selected include a function (straight line), a quadratic function (parabola), an exponential function, a power function, a modified exponential function, and a Gong Baozi curve and a logic curve. Due to limited space, only the most commonly used linear and exponential trend extrapolation methods in market prediction are introduced here.