The fuzzy model of the phase identification on the real estate market

Real estate is considered as an important segment of the financial market in terms of the coherent market hypothesis, which combines the nonlinear deterministic model and statistical dynamic model and its basic assumption is that the dynamics of the market is divided into four phases: random walk, unstable transition, chaotic market and coherent market. Information base for research is statistical information market dynamics on the average price per square meter of real estate in the secondary real estate market, as well as derivative yields a time series of prices. In this article a method for identifying the phases of real estate market, based on the use of fuzzy model. The general scheme of the method consists of three parts: I ? contains fuzzy model identification phase of the real estate market (due to the high level of uncertainty in the market price dynamics and its properties of nonlinearity, nonstationarity, etc.); II ? verification of the model based on the comparison of the results of the identification data of peer reviews; III ? is designed to train the model (optimizing of parameters of membership functions, based on which is carried out the classification). To build a model of the choice of input (explanatory) factors model, which involves the separation of the most informative features for the classification phase. For this goal, the properties of the dynamics of return rates in the phases of the coherent market requirements regarding quantitative indicators were analysed. As a result, quantitative characteristics of dynamics were selected, which were obtained by the use of complex statistical and fractal analysis and characterize the nature of the time series, estimate the depth of its memory, the local dynamics of resistance and the presence of landslides. The knowledge base model is built. The application of the constructed model will allow to implement a qualitative analysis of the current situation in the real estate market, to provide guidance on the selection of relevant instruments prediction.