节点文献
数列修匀的加权算术平均数──中位数混合移动平均法
A New Graduated Method: the Mixed Moving Average Method of Weighted-Arithmetic-mean and Median
【摘要】 在研究变量数列修匀技术的光滑性和拟合度准则的基础上,将中位数引入到加权移动平均法公式中,构建新的数列修匀方法:加权算术平均数──中位数混合移动平均法.论文的后半部分通过实例对加权移动平均法、加权算术平均数──中位数混合移动平均法、指数平滑法和Bayesian方法等四种修匀方法的光滑性和拟合度作了对比研究,最后得出结论:加权算术平均数中位数混合移动平均法具备了计算方法简便、光滑性能好、拟合度较高等特点,对变量数列具有广泛的适应性和良好的修匀效果.
【Abstract】 Through the analysis of two properties of smoothing and fitting, this paper puts forward a new graduated method: the mixed moving average method of weighted-arithmetic-mean and median. Compare four methods of Weighted-Moving-Average, Exponential-Smoothing, Bayesian Graduation and the New Graduated Method, the new graduated method would generate a new series of varieties with a good smoothing and excel- lent fitting.
【关键词】 数列修匀;
光滑性;
拟合度;
中位数;
混合移动平均;
【Key words】 graduation of series; smoothing; fitting; median; mixed moving average method;
【Key words】 graduation of series; smoothing; fitting; median; mixed moving average method;
- 【文献出处】 昆明理工大学学报(自然科学版) ,Journal of Kunming University of Science and Technology , 编辑部邮箱 ,2001年05期
- 【分类号】F224
- 【被引频次】6
- 【下载频次】270