http://sedici.unlp.edu.ar:80/handle/10915/17792 2013-05-19T17:32:05Z 2013-05-19T17:32:05Z Sosa Escudero, Walter http://sedici.unlp.edu.ar:80/handle/10915/3500 2012-06-27T19:27:09Z 2001-01-01T00:00:00Z

Documento de trabajo Documentos de Trabajo; no. 29 Even though the result recently referred to as the "Frisch-Waugh-Lovell theorem" (FWL theorem, henceforth) has been around for a long time, it is relatively recently that it has been widely used by econometricians as a powerful pedagogical tool to express in a simple and intuitive way many results that often rely on tedious and seldom intuitive algebraic steps, which are also notationally cumbersome. Even though a proof of the FWL theorem can be based entirely on standard algebraic results, the main reason of its increasing popularity is its strong geometric appeal. Recent texts and articles provide a mix between algebraic proofs and geometrical illustrations of the theorem, but none of them presents a fully geometrical proof of the result. The goal of this note is very modest: it extends the standard geometrical representations of the theorem to actually prove it based on geometrical arguments, which should, hopefully, provide a richer understanding of the scope of the theorem.

2001-01-01T00:00:00Z Even though the result recently referred to as the "Frisch-Waugh-Lovell theorem" (FWL theorem, henceforth) has been around for a long time, it is relatively recently that it has been widely used by econometricians as a powerful pedagogical tool to express in a simple and intuitive way many results that often rely on tedious and seldom intuitive algebraic steps, which are also notationally cumbersome. Even though a proof of the FWL theorem can be based entirely on standard algebraic results, the main reason of its increasing popularity is its strong geometric appeal. Recent texts and articles provide a mix between algebraic proofs and geometrical illustrations of the theorem, but none of them presents a fully geometrical proof of the result. The goal of this note is very modest: it extends the standard geometrical representations of the theorem to actually prove it based on geometrical arguments, which should, hopefully, provide a richer understanding of the scope of the theorem.