Using robust outlier detection to identify possible flood events

  • Authors:

  • Azami Zaharim;Rafizah Rajali;Kamarulzaman Ibrahim

  • Affiliations:

  • Fundamental Engineering Unit Faculty Engineering and Built Environment, Universiti Kebangsaan Malaysia, Selangor, Malaysia;Department of Statistics, Universiti Kebangsaan Malaysia, Selangor, Malaysia;Department of Statistics, Universiti Kebangsaan Malaysia, Selangor, Malaysia

  • Venue:
  • ICOSSSE'08 Proceedings of the 7th WSEAS international conference on System science and simulation in engineering
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

The least squares method has been widely used in time series forecasting and outlier detection. However, the method is not very efficient in identifying outliers because it suffers the masking effect. The aim of this study is to overcome the masking effect by implementing the robust least median squares method in outlier detection. To illustrate, we identified the possible outliers from sixty-one readings of the daily rainfall recorded at Kajang JPS telemetric station. The outliers are then categorized into innovational outlier and additive outlier. Results based on both techniques were compared and it is found that the least median squares method effectively unmasked the effect of outliers as compared to the least squares method.