عنوان مقاله [English]
The main hypothesis of this study is that, the economic growth can't be a linear and continues process at least for some countries. Also we are interested to factors which drive growth on nonlinear and occasionally discontinues path. We apply catastrophe, chaos and neural networks to show the nonlinearity and jumps in the economic growth path.
Nonlinearity and sudden jumps is confirmed by empirical results of the catastrophe models for NICs economic growth. In addition to more than 500 regressions with different control variables in each regression for individual countries, some panel regression for all countries data confirms the validity of individual results. Chaos models in logistic form results in negative Lyapunov exponent but not significant statistically which can’t be reliable. However, Lyapunov exponent estimation by means of Matlab gives negative exponent that are reliable for some of the countries. Terasvirta and White nonlinearity test reject linearity in growth path for Singapore and Thailand. Also jump test agree with our hypothesis which involve with existence of sudden jumps in growth path for all of NICs.
Results of this study imply policy prescription as follows: nonlinear targeting in growth, focus on foreign resources for enhancing growth potentials and making institutions performance better to promote growth rate.