Description: This book describes the development between 1750 and 1900 of what we would now regard as a class of statistical fitting procedures. In their early stages, these procedures had very little statistical content, and we therefore refer to the subject area by its historical name of the calculus of observations. The book contains detailed algebraic descriptions of the fitting of linear relationships by the method of least squares and the closely related least absolute deviations and minimax absolute deviations procedures. These traditional fitting procedures are of course addressed in conventional statistical textbooks, but in such books the discussion of their historical background is usually extremely slight if not entirely absent. The primary audience for this book will be statisticians concerned with the fitting of linear models. However, it will also be of interest to engineers and scientists concerned with the empirical determination of linear relationships.
Review Quotes:
From the reviews:
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
"Farebrother gives the details of the algebraic development of various procedures, striving to make them more understandable by modernizing and standardizing the notation as much as possible without departing too far from that used by the original developers. ...Readers interested in the history...will be gratified by the helpful historical and bigraphical notes...I thoroughly enjoyed reading this book, and recommend it without hesitation to readers interested in the subject."