DocumentCode
2727071
Title
On nonparametric curve estimation with compressed data
Author
Pawlak, M. ; Stadtmüller, U.
Author_Institution
Dept. of Electr. & Comput. Eng., Manitoba Univ., Winnipeg, Man., Canada
fYear
1995
fDate
17-22 Sep 1995
Firstpage
253
Abstract
Modified kernel estimators calculated from compressed data for density estimation and signal recovery problems are proposed. An asymptotically optimal compression technique utilizing the quantile process and data binning is employed. The statistical accuracy of the introduced kernel estimators is studied, i.e., we derive mean squared error results for the closeness of the these estimators to both the true functions and the kernel estimators determined from non-compressed data
Keywords
data compression; density; estimation theory; optimisation; signal reconstruction; asymptotically optimal compression technique; compressed data; data binning; density estimation; kernel estimators; mean squared error results; modified kernel estimators; noncompressed data; nonparametric curve estimation; quantile process; signal recovery problems; statistical accuracy; Computational complexity; Convergence; Density functional theory; Information theory; Kernel; Random variables; System identification;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory, 1995. Proceedings., 1995 IEEE International Symposium on
Conference_Location
Whistler, BC
Print_ISBN
0-7803-2453-6
Type
conf
DOI
10.1109/ISIT.1995.535768
Filename
535768
Link To Document