Forecasting the Iranian Tax Revenues: An Application of Nonlinear Models

Taxes are one of the main components of the government budgetand as such, having a clear image of the attainable amount of taxes is not only necessary for optimal allocation of scare resources for tax collection purposes, but also helps the government to develop more precise tax collection plans. In this article, the structural features of taxrevenues series have first been examined in terms of linearity, chaotic nonlinearity and stochasticity using Lyapunov Exponent. These series are: total taxes, direct taxes, indirect taxes, corporate taxes, income taxes, salary taxes, real estates taxes, business taxes, wealth taxes, inheritance taxes and goods & services taxes. The results endorse the existence of chaos in the series of different tax resources in different extents. Therefore, based on the results obtained, it was found that more accurate short-term forecasts are possible by applying nonlinear modeling methods. In the next step, using data of the period 1963- 2006, tax revenues of different resources were forecasted for the period 2007-2009 by applying both parallel and proposed multipleinput multiple-output structures of the ANN’s.