Provide a step by step procedure for the application of the Generalized Extreme Studentized Deviate Many-Outlier Procedure to a set of test results that is approximately normally distributed.
Conventional outlier identification techniques such as Grubbs, Dixon suffer from masking. This GESD technique does not suffer from masking, and has been recommended by CS94 to the National Exchange Group and other round robin testing groups as a reliable and robust technique for the simultaneous identification of multiple outliers in a data set that is expected to be approximately normally distributed. This technique has been successfully deployed to identify multiple outliers in round robin test data by the NEG and other exchange groups. The technique has been described and filed with ASTM as Research Report D02-1481 in 2000. The ASTM ILCP program has recently made the decision to adopt this GESD technique for outlier identification. Hence, Subcommittee 94 has decided to develop this Standard Practice to formally adopt the GESD technique into ASTM D02.94 family of standards.
The title and scope are in draft form and are under development within this ASTM Committee.Back to Top
At Society Review