Comparison Between Classical Methods with Spectral Analysis to Obtain the Adaptive Smoothing Constant in the Presence of Outliers

Authors

• Tahir Reisan Dikheel, Department of Statistics, College of Administration and Economics, University of Al-Qadisiyah, Diwaniyah, Iraq
• Zainab yousif Dawood, Department of Statistics, College of Administration and Economics, Mustansiriyah University, Iraq

Abstract

In this paper , two method which deal with finding the optimal value for adaptive smoothing constant, are compared .This constant is used in adaptive Single Exponential Smoothing (ASES). The comparing is between a method uses time domain and another uses frequency domain when the data contain outlier value for autoregressive model of order one AR(1) , or Markov Model, when the time series are stationary and non stationary with deferent samples.

DOI

10.33095/jeas.v13i45.1157

Subject Area

Statistical

First Page

304

Last Page

314

Recommended Citation

Dikheel, T. R., & Dawood, Z. y. (2007). Comparison Between Classical Methods with Spectral Analysis to Obtain the Adaptive Smoothing Constant in the Presence of Outliers. Journal of Economics and Administrative Sciences, 13(45), 304-314. https://doi.org/10.33095/jeas.v13i45.1157