Iranian Journal of Economic Research
Year:2002 | Volume:4 | Issue:12|Papers Count:6

The objective of this paper is to investigate the dynamics of adjustment to long-run purchasing power parity (PPP)in a nonlinear framework using the Iranian data over the period 1959-2000. Two PPP measures are considered and nonlinearity in the real exchange rates are investigated. First, the standard and modified unit root tests are applied and then, cointegration analysis is carried out, based on the Johansen (1988) and Johansen and Juselius (1990) cointegration methodology, rather than imposing the strict cointegating vector in calculating real exchange rate measures. Furthermore, the Smooth transition autoregressive (STAR) representation for the adjustment process towards PPP, which provides a superior alternative, are specified and estimated. It was found that purchasing power parity (PPP) holds in the long-run after accounting for structural breaks. Moreover, linearity was strongly rejected and the dynamic STAR models were specified and estimated by using nonlinear least squares. The strong evidence in favour of nonlinear behaviour for PPP suggests that the linear models are misspecified.

Sudden jumps, qualitative changes and discontinuities are not rare in social and economic phenomena. Catastrophe theory is a proper approach to modeling of dynamical systems that a company faces with sudden jump. Catastrophe theory can be applied in areas such as nonlinear growth models, technical changes, institutional changes, Philips curve, stock market, consumers' behavior and monopolistic behavior. This paper tends to clarify the theory and provide a general guideline to application of it in specific subjects.

Chaos theory is rather new in science, but, it is, in fact, rooted in ancients' perception of the world. The main idea is that although a complex system, such as world, seems to be generated by a random, and therefore, unpredictable process, it may run by a nonlinear deterministic process. Chaos theory has been applied to some Economic time series to see if they have an order, and, therefore, predictable. Some Economic time series, such as stock prices, look random, but, according to the chaos theory, they may come from a nonlinear deterministic process. If the data generating process is nonlinear, using traditional linear methods in estimation and forecasting can be misleading. Chaos theory is also applied to macroeconomic models. Some macroeconomic concepts, such as endogenous business cycles, can now be explained by the theory. In this paper, I try to review the chaos theory and its mathematical root for economists. Then, I will survey the Economic applications of the theory, and finally, will analyze different methods introduced for testing for chaos.

Application of non-classical methods in modeling complex systems and forecasting their behavior has become as more as usual for the scientists and professionals. In most complex systems, especially in non-linear systems, application of classical methods is very difficult or even useless. Non-classical methods are intelligent, knowledge-based, very flexible, and therefore effective in modeling and forecasting. Neural Networks are one of the well-known and innovative nonclassical methods, which have being used in modeling, pattern recognition, clustering and forecasting. This paper tries to predict the economic time series by neural nets. Economic time series are considered as outputs of complex and non-linear economic systems, which can be modeled and forecasted by the neural nets.It has been shown that the performance of neural nets (as prediction machine) is very sensitive and dependent on the structure, size, and learning method of neural nets. In this paper, using MATLAB neural nets toolbox, some Iranian economic time series are being used as case studies for the neural network application.

Artificial neural networks(ANN) are flexible models used for data analyzing and Modeling non-linear relations.Most economic applications of the ANN models have been in financial markets. Only recently there have been same macroeconomic applications of ANN models. In this paper, we set up an ANN model to forecast Iranian Economic growth using a long data from 1315 to 1375. We used different inputs based on Economic growth models and time series forecasting models in the ANN model. The forecasting results are then compared with those of the economic and time series models. The results show that in most of the cases, the ANN model outperform other traditional forecasting models.