Bayesian statistical analysis of censored data in geotechnical engineering

The geotechnical engineer is often faced with the problem ofhow to assess the statistical properties of a soil parameter on the basis ofa sample measured in-situ or in the laboratory with the defect that somevalues have been replaced by interval bounds because the corresponding soilparameter values have turned out to exceed the capacity of the measuringinstrument.

Given such a censored sample the problem is to estimate themean value, variance and certain fractiles of the distribution of the soil pa-rameter as it would be measured by an instrument of unrestricted range. Inmany situations only a small censored sample is given. To estimate the char-acteristic value de¯ned as a lower fractile value corresponding to a codi¯edprobability value, the geotechnical engineer is thus urged to supplement theinformation from the measurements at the actual location by consideringwhatever prior knowledge is available about the soil parameter distribution.The present paper shows how a characteristic value by computer calcula-tions can be assessed systematically from the actual sample of censored datasupplemented with prior information from a soil parameter data base.