PERFORMENCE OF FORECASTING MODEL FOR AVERAGE SURFACE AIR TEMPERATURE PREDICTION OVER SOUTHEAST ASIA

Abstract

In this study, the problem of prediction of a given time series has been investigated. The monthly experimental data sets of average surface air temperature over Southeast Asia are analyzed using Educational Global Climate Model (EdGCM). An attempt is being made to examine the performance of a forecasting model based on concept of error growing rate measurement namely Lyapunov exponent (LE). Here, LE and Supremum Lyapunov exponent (SLE) have been evaluated for average surface air temperature under climate change condition. From the LE and SLE behavior, it showed a clear decrease in error growing rate of average surface air temperature after 90-years forecast. This results suggest general good performance for surface air temperature prediction method.