Inflation forecasting by hybrid singular spectrum analysis – multilayer perceptrons neural network method, case of Indonesia

Fajar, Muhammad and Hartini, Sri (2017): Inflation forecasting by hybrid singular spectrum analysis – multilayer perceptrons neural network method, case of Indonesia. Forthcoming in:

Abstract

Inflation is one of the most important macroeconomic indicators which affects the economic condition of a nation. Therefore, it is necessary to maintain its stability in order that it will not lead to a negative impact and an economic vulnerability. The drastic change in the rate of inflation is determined by the condition of the price of goods which is affected by the distribution and supply-demand factors of goods. As a consequence, it becomes a very important act of action to control inflation. This can be achieved by meeting the information needs of future inflation rates that is needed for the government and the policy of the monetary authority. Fulfillment of accurate and reliable future forecasts of future inflation estimates can be obtained through forecasting. This paper examines the application of the method of Hybrid singular spectrum analysis - a multilayer perceptions neural network to predict the inflation. The main data source used is monthly inflation (in percent) collected by BPS Statistics Indonesia. The result of the study found that the ability of SSA-MPNN Hybrid method is good enough in predicting monthly inflation, as it is provided by the MAPE value of 35.42 percent, without-sample of three observations.

• Item Type: MPRA Paper
• Language: English
• Keywords: inflation, forecasting, hybrid singular spectrum analysis-multilayer perceptions neutral network, Indonesia
• Subjects: C - Mathematical and Quantitative Methods > C2 - Single Equation Models ; Single Variables > C22 - Time-Series Models ; Dynamic Quantile Regressions ; Dynamic Treatment Effect Models ; Diffusion Processes; C - Mathematical and Quantitative Methods > C4 - Econometric and Statistical Methods: Special Topics > C45 - Neural Networks and Related Topics; C - Mathematical and Quantitative Methods > C5 - Econometric Modeling > C53 - Forecasting and Prediction Methods ; Simulation Methods; E - Macroeconomics and Monetary Economics > E1 - General Aggregative Models > E17 - Forecasting and Simulation: Models and Applications

References

[1] Aladag, C.H., Egrioglu, E, & Kadilar, C. ,2012, Improvement in forecasting accuracy using the Hybrid Model of ARFIMA and Feed Forward Neural Network. American Journal of Intelligent Systems 2 (2), p 12-17.

[2] Alexandrov, Th., Golyandina, N. , 2005, Automatic extraction and forecast of time series cyclic components within the framework of SSA. In Proceedings of the 5th St.Petersburg Workshop on Simulation, June 26 – July 2, 2005, St.Petersburg State University, St.Petersburg: pp. 45–50.

[3] Bank Indonesia [Central Bank of Indonesia], 2012, Koordinasi Pengendalian Inflasi [Coordination of Inflation Control], dated 18th February 2018, can be downloaded from http://www.bi.go.id/id/moneter/koordinasi-pengendalian-inflasi/Contents/Default.aspx

[4] Chang, P-C., Wang, Y-W., & Liu, C-H., 2007, The development of a weighted evolving fuzzy neural network for PCB sales forecasting. Expert Systems with Applications 32, p 86–96.

[5] Cryer, J.D., & Chan, K.S., 2008, Time series analysis: with application in R. Second Edition. USA: Spinger Science and Businiess Media, LLC.

[6] Fajar, M., 2016, Perbandingan Kinerja Peramalan Pertumbuhan Ekonomi Indonesia Antara ARMA, FFNN dan Hybrid ARMA-FFNN. [Performance comparison of Indonesia economic growth-forecast among ARMA, FFNN and Hybrid ARMA-FFNN]. DOI: 10.13140/RG.2.2.34924.36483

[7] Fajar, M., 2018, Meningkatkan Akurasi Peramalan dengan Menggunakan Metode Hybrid Singular Spectrum Analysis-Multilayer Perceptron Neural Networks [Improving the forecasting accuracy by Hybrid Method Singular Spectrum Analysis-Multylayer Perceptron Neural Networks]. DOI: 10.13140/RG.2.2.32839.60320.

[8] Kourentzes,N, Barrow, D. K., & Crone, S.F., 2014, Neural network ensemble operators for time series forecasting. Expert Systems with Applications 41, p 4235–4244.

[9] Makridakis, W., & Mac Gee, 1999, Metode dan Aplikasi Peramalan [Method and Application on Forecasting]. Jakarta, Binarupa Aksara.

[10] Rahmani, Donya., 2014, A forecasting algorithm for Singular Spectrum Analysis based on bootstrap Linear Recurrent Formula coefficients. International Journal of Energy and Statistics, Vol. 2 (4), p 287–299.

[11] Wei, W. W. S., 2006, Time series Analysis: Univariate and Multivariate Methods. California: Pearson Education, Inc.

[12] Zhang, G.P., 2003, Time series Forecasting using a Hybrid ARIMA and Neural networks Model. Neurocomputing, 50, p 159.

[13] Zhang, Q., Wang, B.D.,He,B., Peng, Y., & Ren, M.L., 2011, Singular Spectrum Analysis and ARIMA Hybrid Model for Annual Runoff Forecasting. Water Resour Manage, 25, p 2683-2703.