Introduction to Nonparametric Estimation

Developed from lecture notes and ready to be used for a course on the graduate level, this concise text aims to introduce the fundamental concepts of nonparametric estimation theory while maintaining the exposition suitable for a first approach in the field.

---

This is a revised and extended version of the French book. The main changes are in Chapter 1 where the former Section 1. 3 is removed and the rest of the material is substantially revised. Sections 1. 2. 4, 1. 3, 1. 9, and 2. 7. 3 are new. Each chapter now has the bibliographic notes and contains the exercises section. I would like to thank Cristina Butucea, Alexander Goldenshluger, Stephan Huckenmann, Yuri Ingster, Iain Johnstone, Vladimir Koltchinskii, Alexander Korostelev, Oleg Lepski, Karim Lounici, Axel Munk, Boaz Nadler, AlexanderNazin,PhilippeRigollet,AngelikaRohde,andJonWellnerfortheir valuable remarks that helped to improve the text. I am grateful to Centre de Recherche en Economie et Statistique (CREST) and to Isaac Newton Ins- tute for Mathematical Sciences which provided an excellent environment for ?nishing the work on the book. My thanks also go to Vladimir Zaiats for his highly competent translation of the French original into English and to John Kimmel for being a very supportive and patient editor. Alexandre Tsybakov Paris, June 2008 Preface to the French Edition The tradition of considering the problem of statistical estimation as that of estimation of a ?nite number of parameters goes back to Fisher.
However, parametric models provide only an approximation, often imprecise, of the - derlying statistical structure. Statistical models that explain the data in a more consistent way are often more complex: Unknown elements in these models are, in general, some functions having certain properties of smoo- ness.