Detection of multiple gross errors based on variance inflation model
Gui Qing-ming, Gong Yi-song, Guozhong Li · Cehui kexue · 2006
Ascribing gross errors to random model,this paper puts forward a detection method of multiple gross errors based on variance inflation model.Taking surveying adjustment practice into account,we first give the score test statistics on single gross error under the condition of unequal weight and individual observations.After that,we present two schemes based on variance inflation model to locate more than one gross error.Finally,as an example we compute and analyze a side-angel network.Many experiments show that the method given is feasible.This resolution is simple and the result is perfect comparative.