ABSTRACT
This chapter illustrates the algorithms with an example and a few remarks. The minimum sum of absolute errors (MSAE) regression is considered a robust alternative to least squares regression by a number of authors. In fact, it is useful to note that the MSAE regression is to the least squares regression what the sample median is to the sample mean. In regression analysis and, in fact, in most statistical analyses, it is tacitly assumed that all the observations have been taken accurately and precisely. The quality of the remaining observations is never questioned or checked. It is possible that the remaining observations are off because of the bias in the instruments or because the observations were taken on different days, at different places, or by different individuals. The chapter examines critically a data set and the MSAE regression fit and thus has more confidence in the results and conclusions.
