The equivalent weight function in robust estimation is usually constructed by the statistics of normal distribution.
抗差估计等价权函数一般由正态分布统计量构造,其临界值(或称准则)一般由实际经验确定。
Critical values of the equivalent weight function are often fixed in constants in robust estimation.
抗差估计等价权函数中临界值一般定为常数。
The advantage and disadvantage of the existing equivalent weight functions are analyzed.
首先分析了现有几种等价权函数的优缺点,然后以适合于测量领域计算的IGGⅢ方案为基础,利用服从于t分布的学生化残差构造等价权函数来进行抗差估计。