A study of Real Estate Bubble Models

In recent years, real estate has boomed in most of the world, and in many countries, real estate prices were raised abnormally, whether real estate market is a bubble has becomes a key issue for both governments and contractors.
The purpose of this dissertation is to develop models to explore the characteristics of real estate bubbles in the last twenty-six years in Taiwan.
This research develops decision models for evaluating a contractors’ risk in handling real estate bubbles. By employing multivariable analysis and CUSUM charts, including both macroeconomic and microeconomic measure. We find that there are 12 macroeconomic principle variables which influence the results of real estate bubbles. And there are 28 microeconomic principle variables which influence the results of real estate. In addition, 23 samples are taken from the construction corporations in Taiwan. The statistical analysis from 6 failure companies and 17 healthy companies is used to identify the significance of financial variables and overcoming the static analysis shortcomings of traditional precautionary models. Then, using the time series analysis of vector Auto-Regressive Moving-Average model (VARMA), it is shown that it is possible to integrate VARAM with the multivariable CUSUM model.
The tests of models verify that models are able to interpret the real situation of the contractors’ risk handling decision. In addition, because the CUSUM chart could provide the “Real-time function,” the precision and practicality of the predicting model is increased significantly. The model could be a reference for the financial adjusting strategies in domestic construction companies and further assist governmental policies to regulate and manage the construction industry.