Subscribe Now Subscribe Today
Science Alert
 
FOLLOW US:     Facebook     Twitter
Blue
   
Curve Top
Asian Journal of Scientific Research
  Year: 2008 | Volume: 1 | Issue: 2 | Page No.: 130-137
DOI: 10.3923/ajsr.2008.130.137
Detection of Outliers in Time Series Data: A Frequency Domain Approach
O.I. Shittu and D.K. Shangodoyin

Abstract:
We consider the identification and detection of outliers in frequency domain using the spectral method. By assuming both the additive and multiplicative effect of outliers on a series, the parameters of the model were estimated using the maximum likelihood method with a view to measuring the effect of the suspected outlier on the parameter of the series. The occurrence of outliers has led to a shift in the phase and amplitude of the Fourier series thus affected the periodogram estimates. Further more, detection of aberrant observations is more exact in the frequency domain than in the time domain.
 [Fulltext PDF]   [Fulltext HTML]   [XML: Abstract + References]   [References]   [View Citation]  [Report Citation]
 RELATED ARTICLES:
  •    Controller Design for an Uncertain Contactless Power Transfer System
  •    Estimation of P (Y<X) in the Rayleigh Distribution in the Presence of k Outliers
How to cite this article:

O.I. Shittu and D.K. Shangodoyin, 2008. Detection of Outliers in Time Series Data: A Frequency Domain Approach. Asian Journal of Scientific Research, 1: 130-137.

DOI: 10.3923/ajsr.2008.130.137

URL: https://scialert.net/abstract/?doi=ajsr.2008.130.137

 
COMMENT ON THIS PAPER
 
 
 

 

 
 
 
 
 
 
 
 
 

 
 
 
 
 
 
 
 

Curve Bottom