Time Series Analysis of Global Average Absolute Sea Level Change Using Nonlinear Autoregressive Neural Network Model

Abstract

This study tried to pursue analysis of time series data using long-term records of global average absolute sea level change from 1880 to 2014 extracted from the U.S. Environmental Protection Agency using data from Commonwealth Scientific and Industrial Research Organization. Using the LM algorithm, the results revealed that the nonlinear autoregressive neural network model with 7 neurons in the hidden layer and 7 time delays provided the best performance at its smaller MSE value. The findings in this study may be able to bridge an important gap in time series forecasting by combining the best statistical and machine learning methods. In order to sustain these observations, research programs utilizing the resulting data should be able to significantly improve our understanding and narrow projections of future sea level rise and variability.