摘要
提出了两个正态分布的混合分布参数估计的新方法,包括矩估计法和信息矩估计法。后者可根据前馈信息的不同而分成3种。为获得更好的符合度,建议将现有的各种估计方法与交互反馈最小卡方估计程序相结合。为同一目的,进一步提出了称为蒙特卡罗最小卡方估计法的估计程序。用假设混合分布的模拟样本对这些方法的正确性作了检测,并对现有方法进行了分类。
New methods for the estimation of parameters in a mixture of two normal distributions were proposed,including the Moment Method of Estimation and the Information-Moment Method of Estimation,which could be distinguished into three kinds according to the difference in feedforward information.In order to get better fitness,it was suggested that each existing method of estimation be combined with the procedure of Interfeed- back-Minimun Chi-square Estimation.For the same purpose,a new proce- dure called the Monte Chrlo-Minimun Chi-square Method of Estimation was proposed further.The correctness of these methods were examined with a simulated sample of a hypothetic mixed distribution.The existing mehods were classified.
关键词
农业数学
混合分布
分布参数
估计
mixed distribution
distirbution parameter
estimation method
