ECONOMIC MODELLING, vol.29, no.4, pp.1161-1169, 2012 (SSCI)
This paper considers movements of Foreign Direct Investments (FDI) in Turkey, and therefore, to understand the dynamics of FDI, runs several nonlinear FDI equations in which the basic determinants of FDI in Turkey are determined through Markov Regime-Switching Models (MSMs). The statistical properties of Markov Regime-Switching time series models are more desirable than those of conventional time series or panel regression models. Through these properties of MSMs, i) one can observe structural changes, if they exist, in FDI equations through time, ii) if, in fact, the true FDI regression equation follows a nonlinear relationship, MSMs fit data better than the linear models. This paper eventually follows maximum likelihood methodology of Markov Regime-Switching Model (MSM) to search for the possible structural changes in level and/or trends and possible changes in parameters of independent variables of FDI-MSM equations through the transition probabilities. In conclusion, this paper yields the outcome that Turkish FDI growth equation has significant structural changes in level and trend and that has significant coefficient shifts in explanatory variables. These explanatory variables are Turkish GDP Growth, Labor Cost, the Electricity Price Growth, the growth in average prices of High Sulphur Fuel Oil, Cooking Coal, Steam Coal and Natural Gas, Export Growth, Import Growth, Discount Rate and Country Risk Indexes for Turkey. US and EU, respectively, within the time interval from 1988 first quarter to 2010 second quarter. (c) 2012 Elsevier B.V. All rights reserved.