TY - JOUR
T1 - Wavelet transform asymmetric winsorized mean in detecting outlier values
AU - Al-Khazaleh, Ahmad M.H.
AU - Al Wadi, S.
AU - Ababneh, Faisal
N1 - Publisher Copyright:
© 2015 Pushpa Publishing House, Allahabad, India.
PY - 2015
Y1 - 2015
N2 - 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.
AB - 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.
KW - Amman Stock Exchange
KW - Asymmetric Winsorized mean
KW - Detecting outliers
KW - Wavelet transform
UR - http://www.scopus.com/inward/record.url?scp=84929175204&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84929175204&partnerID=8YFLogxK
U2 - 10.17654/FJMSFeb2015_339_351
DO - 10.17654/FJMSFeb2015_339_351
M3 - Article
AN - SCOPUS:84929175204
SN - 0972-0871
VL - 96
SP - 339
EP - 351
JO - Far East Journal of Mathematical Sciences
JF - Far East Journal of Mathematical Sciences
IS - 3
ER -