Authors 
: 

Document type 
: 
Communication à un colloque (Conference Paper) – Présentation orale avec comité de sélection

Abstract 
: 
Given random variables X isin R/sup d/ and Y such that E[YX=x]=m(x), the average derivative delta /sub 0/ is defined as delta /sub 0/=E[ nabla m(X)], i.e., as the expected value of the gradient of the regression function. Average derivative estimation has several applications in econometric theory (T.M. Stoker, 1992) and thus it is crucial to have a fast implementation of this estimator for practical purposes. We present such an implementation for a variation known as density weighted average derivative estimation. This algorithm is based on the ideas of binning or Weighted Averaging of Rounded Points (WARPing). The basic idea of this method is to discretize the original data into a d variate histogram and to replace in the nonparametric smoothing steps the actual observations by the appropriate bincenters. The nonparametric smoothing steps become thus a (multidimensional) convolution between the (discretized) data and the (discretized) smoothing kernel. A Monte Carlo study demonstrates that with this binned implementation substantial reduction in computing time can be achieved. But it will also become clear that in higher dimension the choice of how to bin is crucial. 
Publication date 
: 
1995 
Conference 
: 
"Proceedings of 26th Symposium on the Interface of Computing Science and Statistics (INTERFACE '94)", Research Triangle Park, NC, USA (1518 June 1994) 
Peer reviewed 
: 
yes 
Host document 
: 
Sall, J.; Lehman, A.; ; "Computing Science and Statistics. Vol.26. Computationally IntensiveStatistical Methods. Proceedings of the 26th Symposium on the Interface" p. 2833 
Publisher 
: 
Interface found. north america 
Affiliation 
: 
UCL 
Links 
: 
