Lazslo Giorfy – A Distribuiton-Free Theory of Nonparametric Regression

There is a long history of the regression estimation problem. In 1632 Galileo Galilei used a procedure which can be seen as a linear relationship to contaminated observed data. The classical linear regression problem involvestting a line through a cloud of points. The principle of least squares was discovered by A. M. Legendre and C. F. Gauss and was published in 1805 and 1809. The principle of least squares can be used to construct nonparametric regression estimates, where one does not restrict the class of possible relationships, and will be one of the approaches studied in this book. The concept of a regression function was introduced by F. Galton in 1889, while a probabilistic approach was already given by A. B- vais in 1846. J. W. Tukey proposed the nonparametric regression estimate in 1947. The partitioning regression can be seen as a special least squares estimate.