DocumentCode :
927303
Title :
The two-to-one rule in data smoothing (Corresp.)
Author :
Papoulis, A.
Volume :
23
Issue :
5
fYear :
1977
fDate :
9/1/1977 12:00:00 AM
Firstpage :
631
Lastpage :
633
Abstract :
If a signal is estimated by a weighted average of the data in the interval 
, then the variance 
of the estimate decreases, but its bias b increases, with increasing c. It is shown that in high accuracy estimates, the mean-square error e is minimum if 
is such that 
, regardless of the form 
of the smoothing weight. Furthermore, the resulting 
is minimum if 
is a truncated parabola.
, then the variance of the estimate decreases, but its bias b increases, with increasing c. It is shown that in high accuracy estimates, the mean-square error e is minimum if is such that , regardless of the form of the smoothing weight. Furthermore, the resulting is minimum if is a truncated parabola.Keywords :
Smoothing methods; Calculus; Estimation error; Matched filters; Random processes; Signal analysis; Signal processing; Smoothing methods; TV; White noise; Wiener filter;
fLanguage :
English
Journal_Title :
Information Theory, IEEE Transactions on
Publisher :
ieee
ISSN :
0018-9448
Type :
jour
DOI :
10.1109/TIT.1977.1055764
Filename :
1055764
Link To Document :
