Document Type : Research Paper

Author

• seyed jalal tabatabaei

Assistant professor

Abstract

The present research has investigated the effect of financial development and internet penetration on the export value of 29 provinces of Iran from 2005 to 2017 using the modified standard error estimation method of panel data and the Gaussian process regression machine learning model. By proposing two main hypotheses based on the research literature, the statistical analysis of the hypotheses has been done according to the estimation results obtained from the collected data. The results of the estimation of the modified standard error of the panel data showed that the use of the Internet has a positive contribution to the country's export value in all provinces. Therefore, the first hypothesis based on the effect of financial development variables and internet penetration on the export growth of Iran's provinces is confirmed. The use of the Internet in developed provinces has a greater impact on export growth, leading to a reduction in time and exchange costs to complete the business process. In less developed provinces, the use of the Internet does not have a significant contribution to the growth of exports. Therefore, it can be said that the Internet penetration rate on the export value depends on the level of development of the provinces, and the second hypothesis of the research that the difference in the effect of the Internet penetration rate in different provinces of the country is confirmed. Also, financial development has a significant and main contribution to the growth of exports in the provinces. The results of the machine learning model showed that according to Gaussian process regression, financial development, gross domestic product, population and the Internet are the most important factors in predicting export growth in Iran's provinces.

Keywords

• internet penetration
• machine learning
• gaussian process

Main Subjects

• Public Economics