Identification of housing market trends and factors affecting them: The greater Haifa metropolitan area in Israel as a case study

We analyzed monthly averages of sale prices of 6,198 residential properties in the Greater Haifa Metropolitan Area, recorded between January 1998 and October 2011. We found the statistical significance of different price influencing factors appeared to change during different phases of housing price formation, with the association between housing prices, on the one hand, and credit and housing supply factors, on the other, being especially strong during the price acceleration phase. We also introduced in the analysis 1st and 2nd derivatives of housing price trend lines, estimated using polynomial smoothing. Although "public sentiments" about housing price trends may influence both supply and demand, they cannot always be modeled directly. Our modeling approach, based on price derivative analysis, may help to capture such sentiments, thus improving the robustness of empirical models used as explanatory tools and for forecasting.