Clustering of regional countries in terms of GDP per capita time series trend over the period of 1990-2014 with ICA Technique

Subject Areas : Regional Planning

جعفر Rahmani Shamsi ¹ (دانشجوی دکترای آمار، دانشکده ریاضی، دانشگاه یزد، یزد، ایران)
فاطمه Asayesh ² (کارشناس ارشد علوم اقتصادی، دانشکده اقتصاد و مدیریت، دانشگاه شهید باهنر کرمان، کرمان، ایران)
علی Dolati ³ (دانشیار گروه آمار، دانشکده ریاضی، دانشگاه یزد، یزد، ایران)

Received: 2016-06-17 Accepted : 2017-11-13 Published : 2017-12-22

Keywords: GDP Per Capita, Vision of the Islamic Republic of Iran, Technique of ICA, Time Series Clustering,

Abstract :

The Vision of the Islamic Republic of Iran in horizon of 2026 has been designed for achieving the first place in economy, science and technology between regional countries. This purpose can be achieved by relative improvement in the level of GDP. In this paper, using clustering method,  the position of Iran among regional countries has been evaluated in terms of GDP per capita and trend of this indicator over a 25 year period (between 1990-2014). For a more precise clustering, the Independent component analysis technique has been implemented as a preprocess. The results indicate that regional countries, in terms of GDP per capita in 2014, can be classified into 4 clusters in which Iran is in the same cluster as Jordan, Iraq, Armenia, Georgia, Sudan, Uzbekistan, Syria, Yemen, Kyrgyzstan, Pakistan, Tajikistan, Lebanon, Turkey, Azerbaijan, Turkmenistan and Kazakhstan. The study also shows that the regional countries can be classified into 9 clusters in terms of GDP per capita time series trend between 1990 to 2014. Based on this criterion, Iran shares the same cluster with Pakistan, Saudi Arabia, Bahrain and Turkey. Considering the significant difference between Iran and countries such as Qatar, UAE, Kuwait, Bahrain, Saudi Arabia and Oman in terms of the GDP per capita, Iran should have grown much faster in recent years. However, this has not been observed according to the clustering results. To achieve the goals of vision in the remaining years, a significant positive change in time series trend of GDP per capita should occur. In such a way, the gap between Iran and other countries can be narrowed and Iran can earned the first place in the region by the end of the vision.

References:

 

1.      Afshordi, Mh. and Madani, SM. (2009): Regional Power System Structure in the Middle east (With an Emphasis on the Superior Countries of the Region). The journal of spatial planning. 13(62), pp: 113-142. (in persian)

2.      Chatfield, C. (2016): The analysis of time series: an introduction. CRC press.

3.      Chen, R. B., Chen, Y., & Härdle, W. K. (2014): TVICA—Time varying independent component analysis and its application to financial data, Computational Statistics & Data Analysis, 74, pp: 95-109.

1. Dehghanpour, H. and Zibaie, M. (2015): MENA countries ranked according to environmental sustainability indicators by TOPSIS. The second National Conference on Agriculture and Development. (in persian)


2. Feizpour, MA. and Asayesh, F. (2014): The Development in Iran and Comparison with other Countries of Region; The Vision 2025 Document. Majlis and Rahbord, 78, pp: 85-119. (in persian)


3. Feizpour, MA. and Shahmohammadi, A. (2014): Regional differences in comparative advantage and sustainability indices. Economy and regional development, 8, pp: 1-15. (in persian)


4. Ghasemian, A. (2015): A review of developments in per capita income between 2011-2014. Chamber of Commerce, Industries, Mines and Agriculture Tehran, deputy economic studies. (in persian)


5. Güveli, A. and S. Kılıçkaplan (2000): A Ranking of Their Levels of Socio- Economic Development, Journal of Economic Cooperation 21, pp: 97-114.


6. Hyvärinen, A., Karhunen, J., & Oja, E. (2004): Independent component analysis (Vol. 46). John Wiley & Sons.


7. Johnson, R. A., & Wichern, D. W. (1992): Applied multivariate statistical analysis (Vol. 4). Englewood Cliffs, NJ: Prentice hall.


8. Liao, T. Warren. (2005): .Clustering of time series data—a survey. Pattern recognition 38.11. 1857-1874.


9. Oja, E., Kiviluoto, K., & Malaroiu, S. (2000): Independent component analysis for financial time series, In Adaptive Systems for Signal Processing, Communications, and Control Symposium 2000. AS-SPCC. The IEEE 2000, pp: 111-116, IEEE.


10. Rani, S., & Sikka, G. (2012): Recent techniques of clustering of time series data: a survey. International Journal of Computer Applications, 52(15).


11. Sadeghi Shahdani, M. Meysami, H. Aghajani Memar, E. Ghaemi Asl, M. (2010): Lessons for Vision: Trend analysis of macroeconomic variables of Iran in the recent period, compared to other countries in the region, Rahborde Yas, 22, pp: 102-127. (in persian)


12. Safadi, Th. (2014): Using independent component for clustering of time series data, Applied Mathematics and Computation, 243, pp: 522-527.


13. Worldbank, Economy & Growth Indicator, 1990-2014.


14. Yaghobi Manzari, P. (2010): Comparative economic situation of the region (based on World Bank data). Business reviews, 44.

11.  Khalili iraghi, M. and Masoudi, N. (2007): Check the income gap MENA region, compared to Japan, during the period 1975-2002, according to the theory of convergence. Journal of economic research. 79. (in persian)

17.  Tavakoli, A. and Mahbanouei, B. (2013): Studying the ranking of regional countries competitiveness based on public and private institutions’ indicators and providing some strategies for enhancing the position of Islamic Republic of Iran. Iran’s economic essays. 20, 135-162. (in persian)

20.  Zamanzadeh, H. (2012): Five decades of ups and downs of production: The causes of low growth and destabilize the Iranian economy. Economic news, 136, pp: 102-204. (in persian)