Colloquium

Department of Statistics

Dissertation Proposal
March 14 (Fri) 11 a.m.
Alavi Commons Room, 6625 Everett Tower

Statistical Methodology for Data with Multiple Limits of Detection

Robert Flikkema
Department of Statistics
Western Michigan University

Limitations of instruments used to collect continuous data sometimes leads to obtaining observations lower than a limit of detection. These observations are known as nondetects. They could be zeroes, or positive numbers, but they are too small to be recorded by a measuring device. Nondetects frequently occur in environmental data. Trace amounts of chemicals can exist in soil or groundwater and are undetectable by a machine reading. These observations pose a problem to researchers since the true values are unknown. Imputation methods attempt to "fill-in" the nondetects with values.

Simulations involving constant, random uniform, regression on order statistics, and parameter estimation imputation methods are discussed. This research investigates the performance of each method when estimating summary statistics and recreating data similar to the original distribution for log-normal and gamma data. Samples containing one and two limits of detection are investigated.

All statistics students are expected to attend.

Past colloquiums