Discussion of 'Recent advances in dimension reduction for regression'.
In: 2002 Proceedings of American Statistical Association meeting, 11-15 August 2002, New York.
It is a great pleasure to discuss the cognate papers Cook (2002), Chiaromonte, Cook and Li (2002) and Cook and Li (2002) presented, in that order, at the WNAR invited paper session ‘Recent Advances in Dimension Reduction for Regression’. For brevity, they are referred to here as [C], [C2L] and [CL] respectively. Following them, the mild conditions required for the central (mean) subspace to exist are assumed. The discussion is in three sections: 1. Appreciation. An appreciation of key features of each paper. 2. Questions. A few questions about (extensions of) each paper, and some related comments. 3. Developments. Some personal suggestions, under current investigation, for further developments in the area.
Actions (login may be required)