AL-Qadisiyah Journal For Administrative and Economic sciences

Regression analysis aims to estimate the relationship between the response variable and explanatory variable. Most of the related studies used the ordinary least squares to achieve this goal. Unfortunately, the ordinary least squares is not suitable to address the unknown data distribution. Further, it is not efficient to evaluate the nonlinear relationship. In this article, we use the support vector regression to build the models of the exchange rate of Iraqi dinner. It is a non-parametric method which is no strict conditions about the data. The results refer to the superiority of the support vector regression method over the ordinary least squares method.

The exchange rate plays an important role in influencing various aspects of the economic structure. The importance of this role reflects a set of basic economic indicators, exports and imports and their impact on the development of the balance of payments, which is used to assess the success and safety of any economic system. Growth and inflation are the main objectives of the exchange rate policy, while most developing countries seek to reduce inflation and improve the balance of payment status as the main objective of exchange rate management policy.