Robust Covariance Estimation Software via Nearest Neighbor Variance Estimation (NNVE)
Software to carry out robust covariance estimation by Nearest Neighbor Variance Estimation (NNVE) [Wang and Raftery (2002, J. Amer. Statist. Ass.)] is now available for R and Splus. In the simulation studies published in JASA, this had mean squared error at least 100 times smaller than that of other leading covariance estimators when the proportion of outliers was high. cov.nnve is now available in the covRobust contributed package at http://cran.r-project.org/src/contrib/PACKAGES.html#covRobust The Splus version is available on the S archive of Statlib http://lib.stat.cmu.edu/S/ under the function name cov.nnve cov.nnve is by Naisyin Wang and Adrian Raftery, with contributions by Chris Fraley. References: Wang, N. and Raftery. A.E. (December 2002). Nearest-neighbor variance estimation (NNVE): Robust covariance estimation via nearest-neighbor cleaning (with Discussion). Journal of the American Statistical Association 97(460): 994-1019. Wang, N. and Raftery. A.E. (2000). Nearest-neighbor variance estimation (NNVE): Robust covariance estimation via nearest-neighbor cleaning. Technical Report no. 368, Department of Statistics, University of Washington. Available at www.stat.washington.edu/www/research/reports ------------------------------------------------------------------- Adrian E. Raftery Professor of Statistics and Sociology Director, Center for Statistics and the Social Sciences University of Washington, Box 354320 Phone: (206) 543-4505 Seattle, WA 98195-4320. FAX: (206) 221-6873 Web: www.stat.washington.edu/raftery; www.csss.washington.edu ------------------------------------------------------------------- . . ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at: . http://jse.stat.ncsu.edu/ . =================================================================
