No abstract

This paper was prepared for presentation at the 1999 SPE Mid-Continent Operations Symposium held in Oklahoma City, Oklahoma, 28-31 March 1999.

No abstract

In this paper a new technique is presented for detecting multiple outliers in regression datasets using genetic algorithms. Each dataset contained known outliers and the genetic algorithm implementation was exceptionally accurate in detecting these outliers in all of the datasets tested.The genetic algorithm is an optimization technique based on various biological principles. It is capable of searching for global optima among a vast number of choices. By intelligent but somewhat random generation of subsets of data, potential sets of outliers are identified by minimizing the residual sum of squares produced by the least squares method. Relative improvements across the outlier sets are then analyzed to determine which sets are indeed outlying.

No abstract