Wavelet transform asymmetric winsorized mean in detecting outlier values

Ahmad M.H. Al-Khazaleh, S. Al Wadi, Faisal Ababneh

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

One of the main problems in large datasets is outlier detection, the outliers are detected using Z-score, box plot method, statistical measures and asymmetric Winsorized mean. This paper has a novel method for detecting the outlier values by combining the asymmetric Winsorized mean with the famous spectral analysis function which is wavelet transform (WT). As a result, after comparing the new technique with the previous mentioned methods using financial data from Amman Stock Exchange (ASE), we have found the wavelet transform asymmetric Winsorized mean (WTAWM) is the best method in outlier detections.

Original languageEnglish
Pages (from-to)339-351
Number of pages13
JournalFar East Journal of Mathematical Sciences
Volume96
Issue number3
DOIs
Publication statusPublished - 2015

Keywords

  • Amman Stock Exchange
  • Asymmetric Winsorized mean
  • Detecting outliers
  • Wavelet transform

ASJC Scopus subject areas

  • Mathematics(all)

Fingerprint Dive into the research topics of 'Wavelet transform asymmetric winsorized mean in detecting outlier values'. Together they form a unique fingerprint.

  • Cite this