房地产行业债券信用利差影响因素分析.doc

上海交通大学硕士学位论文
A STUDY ON THE DETERMINANTS OF THE CREDIT
SPREADS OF BONDS IN REAL ESTATE SECTOR
ABSTRACT
In this paper, we use 116 bonds in real estate sector as data sample to
study the determinants of credit spreads with reference to mature models
introduced in foreign literature. We conduct empirical study on structural
model variables using both cross-sectional and time series data regression
method.
Cross-sectional regression results show that bond rating, sinking fund
provision and financial strength of local government have a significant
impact on the credit spreads of bonds in real estate sector. The explanatory
power of regression model reaches % with all variables significant.
We also conduct studies on firm-level variables including leverage ratio,
quick ratio and collateral. The results shows no significant impact of these
variables which indicates that bond investors pay more attention to the
support of local government and pay less to the firm-level data. It can be
mainly attributed to the lagging disclosure of financial data.
The time series regression results of daily frequency data show that
risk-free rate and changes in yield curve structure have significant impact
on changes of credit spreads. Return and volatility of stock market show
no significance. It can be attributed to the fact that most issuers are not
pany. The explanatory power of the model reaches %
which matches the results in foreign literatures. It indicates that structural
model variables have certain explanatory power but we have not find the
main factors affecting changes of credit spreads.
Due to the lack of data points in monthly frequency data, the
regression result is affected and weakened. In general, we found the impact
II
上海交通大学硕士学位论文
of changes in macroeconomic variables contrast with the results we
expected, probably due to the recent financial crisis and the credit default
event shocks happened recently.
KEY WORDS: Credit spreads,