Summary
In business, industry and government agencies, anticipating future behavior that involves many critical variables for nation wealth creation is vitally important, thus the necessity to make precise decision by the policy makers is really essential. Consequently, an accurate and reliable forecast system is needed to compose such predictions. Accordingly, the aim of this research is to develop a new hybrid model by combining a linear and nonlinear model for forecasting time series data. The proposed model (GRANN_ARIMA) integrates nonlinear Grey Relational Artificial Neural Network (GRANN) and linear ARIMA model, combining new features such as multivariate time series data as well as grey relational analysis to select the appropriate inputs and hybridization succession. To validate the performance of the proposed model, small and large scale data sets are used. The forecasting performance was compared with several models, and these include: individual models (ARIMA, Multiple Regression, Grey Relational Artificial Neural Network), several hybrid models (MARMA, MR_ANN, ARIMA_ANN), and Artificial Neural Network (ANN) trained using Levenberg Marquardt algorithm. The experiments have shown that the proposed model has outperformed other models with 99.5% forecasting accuracy for small-scale data and 99.84% for large-scale data. The empirical results obtained have proved that the GRANN_ARIMA model can provide a better alternative for time series forecasting due to its promising performance and capability in handling time series data for both small and large scale data.
See the full content of this document
Extract
Forecasting Time Series Data Using Hybrid Grey Relational Artificial Neural Network and Auto Regressive Integrated Moving Average Model
1. Introduction
Predicting the future is important for the organization to plan or adopt the necessary policies. Forecasting can assist them to make a better development and decision-making for the country. There are various forecasting techniques available in the academic literature. However, the selection of these techniques normally depends on the availability of data, the quality of available models and some predefined assumptions. According to Makridakis et al. [33], each method is different in terms of accuracy, scope, time horizon and cost. To facilitate an adequate level of forecasting accuracy, the developer has to be responsive to the characteristics of different methods, and determine if a particular method is appropriate for the undertaken situation before embarking its usage in real application. As a result, the choice of a forecasting model is one of the important factors that will influence the forecasting accuracy.Forecasting methods can be broadly divided into two categories: Statistical and Artificial Intelligence (AI) based techniques. Box-Jenkins or Auto Regressive Integrated Moving Average (ARIMA), Multiple Regressions and Exponential Smoothing are examples of statistical methods, whilst AI paradigms include fuzzy inference systems, genetic algorithm, neural networks, machine learning etc. [55]. Statistical methods are usually associated with linear data, while neural networks are usually associated with nonlinear data. Statistical methods have been used successfully in time series forecasting for several decades. As well being simple and easy to interpret, statistical methods also have several limitations. One of the major limitations of statistical methods is it is merely depicted as a linear model, also known as model driven approach. Thus, they have to fit the data with the available data and the prior knowledge about the relationships between the inputs and outputs before modeling is highly desired.Due to the limitations of statistical methods, nonlinear statistical time series models have been proposed, with the aim to improve the forecasting performance of nonlinear systems. These include bilinear model, threshold autoregressive model (TAR), smoothing transition autoregressive model (STAR), autoregressive conditional heteroscedastic model (AR...See the full content of this document
Sponsored links
