ANALYSIS OF AURORAL ELECTROJET MAGNETIC INDICES.

In this paper magnetic indices were extracted from World Data Center (WDC) for geomagnetism for six years (2008 to 2013). The indices extracted are AuroralElectrojet (AE), Auroral Upper (AU), Auroral Lower (AL) and Auroral Oval (AO). The indices daily values measured at 24 hours Universal Time (UT) were averaged to monthly values. Time series statistical analysis was carried out on the monthly averaged data using SPSS version 16.0 with expert modeler, to determine the level, trend and seasonal variations of the indices. From the results obtained after the statistical analysis, it can be seen that the indices have seasonal patterns of which AE, AL and AO as March but AU has July. Also from the values of the coefficient of determination, R², it can also be concluded that AE and AL can be better predicted from the models than AU and AO. From the results obtained, simple seasonal model for AE, AL, and AO, and winter's additive for AU was determined. Using the values of R², it can be seen that the models for AE and AL can be used to predict or make forecast of the behavior of the indices. It was also discovered that level (alpha) has more significant contributions in the behavior of the system than seasonal. [ABSTRACT FROM AUTHOR]