论文部分内容阅读
用于估计马斯京根模型参数的方法很多,但这些方法在数据存在异常值时缺乏抵御异常值影响的抗差性能.推导出一种有限制条件的参数抗差估计算法,通过含有随机误差和异常误差的人工数据和真实数据比较抗差算法与传统最小二乘算法的抗差性.研究表明抗差估计算法能减小异常值对参数估值的影响.
There are many ways to estimate the parameters of Muskingum model, but these methods lack the robustness to resist the influence of outliers when there are outliers in the data.A constrained parameter robust estimation algorithm is derived, which contains random error And the error of the artificial and real data to compare the robustness of the robust algorithm and the traditional least squares algorithm.The research shows that the robust algorithm can reduce the influence of outliers on the parameter estimation.