Premium
Dynamic Analyses for Transmission between Asset Bubble Trends of Accumulated Co‐integration Errors
Author(s) -
Kim YunYeong
Publication year - 2016
Publication title -
asia‐pacific journal of financial studies
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.375
H-Index - 15
eISSN - 2041-6156
pISSN - 2041-9945
DOI - 10.1111/ajfs.12139
Subject(s) - bubble , economic bubble , economics , stock (firearms) , econometrics , cointegration , asset (computer security) , stock market bubble , financial economics , financial asset , capital asset pricing model , monetary economics , stock market , finance , computer science , mechanical engineering , paleontology , computer security , horse , parallel computing , engineering , biology
We analyze how the bubble trends of different asset prices are dynamically transmitted between one another under a co‐integrated VAR model. The vector bubble trends are defined by a random walk trend derived from Beveridge‐Nelson decomposition allowing a co‐integration between multiple asset prices and their fundamentals. This I(1) bubble trend is connected to the no‐arbitrage condition for efficient asset pricing and is composed of Accumulated Cointegration Error (ACE). Under this setup, we may conduct dynamic analyses of the vector ACE bubble trends, which enables us to assess the interrelationships of bubbles of different assets and the dynamic role of bubbles in asset pricing. It may be critical for a preemptive policy reaction to block any possible contagion of an asset price bubble that might be harmful to the economy. Using the monthly data of the U.S. and impulse response analyses, we empirically find that (i) there are ACE bubble trends in both stock and housing prices; (ii) the stock and housing price bubbles are highly correlated with one another; (iii) stock prices are more bubble‐dependent than housing prices; and (iv) the stock bubble affects the housing bubble, but the reverse case does not hold.