A New Unified Deriving Method for Gross Error and Systematic Error Significance Test Statistics
The data snooping based on mean shift model and the adjustment method with additional system parame-ters are currently important methods for handling gross errors and systematic errors.Their function models have the same form,so the significance test statistics can be theoretically derived uniformly.Unlike existing literatures,a new unified deriving method has been proposed.The new method conforms to the general idea of constructing sig-nificance test statistics in mathematical statistics,and has the characteristics of simplicity and easy understanding,enriching existing deriving methods.In addition,the new method also has wider applications,and it can also be used to derive gross error and systematic error significance test statistics for other adjustment methods besides pa-rameter adjustment when needed.
万方数据
维普
