Analysis on the Relationship between U.S. Real Estate Prices - Using Copula Model

• Main Author: 卓逸文
• Other Authors: Kuang-Liang Chang
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/2tt3m8

碩士 === 國立嘉義大學 === 應用經濟學系研究所 === 105 === This thesis uses an AR(p)-GARCH(1, 1) model and several Copula functions include Frank Copula function, Clayton Copula function, Gumbel Copula function and Normal Copula function to study the relationship between real estate prices in California (CA), Florida (FL), Illinois (IL), New York (NY) and Texas (TX). The results show that there is a significant positive correlation between the state real estate prices in the Frank Copula function. While the Clayton Copula function indicates that there is a higher degree of correlation when the two variables are falling. The Gumbel Copula function has a higher degree of correlation when the two variables are rising. The Frank Copula model is the best fit model for CA-TX, FL-TX, IL-TX and NY-TX. The Gumbel Copula model is the best fit model for CA-IL, FL-IL, FL-NY and IL-NY. Then, the Normal Copula model was the best fit model for CA-FL and CA-NY. The results also find that the 2008 economic crisis did have a great impact on the US real estate market, real estate prices plummeted, resulting in the rise in the number of law auction housing, but in 2012 with the economic recovery, the real estate market began to recover, real estate prices tend to stable.