Title :
Nonparametric regression estimation for arbitrary random processes
Author :
Posner, S.E. ; Kulkarni, S.R.
Author_Institution :
Dept. of Electr. Eng., Princeton Univ., NJ, USA
Abstract :
We study nonparametric estimates of E[Yn|Xn] of the form Σi=1n-1 Wni(X1 ...Xn)Yi based on Xn and data {(X i,Yi)}i=1n-1. Our work analyses the case where (Xi) is a completely arbitrary random process. Conditions on the weights are established so that the time-average of the estimation errors converges to zero. One consequence of our work is a recovery and extension of some classical results to stationary processes in separable metric spaces
Keywords :
estimation theory; nonparametric statistics; random processes; sequences; statistical analysis; arbitrary random processes; estimation errors; nonparametric regression estimation; separable metric spaces; stationary processes; time-average; weights; Artificial intelligence; Estimation error; Extraterrestrial measurements; Information theory; Kernel; Random processes; Random variables; Space stations; Statistics;
Conference_Titel :
Information Theory, 1995. Proceedings., 1995 IEEE International Symposium on
Conference_Location :
Whistler, BC
Print_ISBN :
0-7803-2453-6
DOI :
10.1109/ISIT.1995.535766
