Abstract for: Sensitivity Analysis of a Real Estate Price Oscillations Model
In system dynamics methodology, formal output analysis can create a basis for improving the structure, so it is important to determine the variables to which the model is sensitive, by using a formal experimental design method. Moreover, such a research can allow comparing different methods for parametric sensitivity analysis. The purpose of this study is to understand different reactions of real estate prices to changes in different input factors. The study is carried out on a system dynamics model previously developed by the authors on real estate prices in Istanbul. The sensitivity of period, amplitude and mean of price oscillations to the changes in some selected variables of the model are analyzed. Two experimental sensitivity analysis designs, namely fractional factorial design and Latin hypercube sampling are used to measure the sensitivity of the model. The study shows that the factors that turn out to be most significant in the two designs are not the same. But the factors and interactions found significant by both techniques can be safely assumed to be highly influential.