Title :
Minimum complexity regression estimation with weakly dependent observations
Author :
Modha, Dharmendra S. ; Masry, Elias
Author_Institution :
Dept. of Electr. & Comput. Eng., California Univ., San Diego, La Jolla, CA, USA
Abstract :
Given N strongly mixing observations {Xi,Yi} i=1N, we estimate the regression function f*(x)=E[Y1|X1=x], x∈ℜd from a class of neural networks, using certain minimum complexity regression estimation schemes. We establish a rate of convergence for the integrated mean squared error between the proposed regression estimator and f*
Keywords :
convergence of numerical methods; estimation theory; neural nets; statistical analysis; convergence rate; integrated mean squared error; minimum complexity regression estimation; neural networks; regression function; strongly mixing observations; weakly dependent observations; Computer networks; Convergence; Kernel; Neural networks; Random variables;
Conference_Titel :
Information Theory and Statistics, 1994. Proceedings., 1994 IEEE-IMS Workshop on
Conference_Location :
Alexandria, VA
Print_ISBN :
0-7803-2761-6
DOI :
10.1109/WITS.1994.513898
