The Empirical assessment of Portfolio Balance Model
Using asset prices to explain the fluctuations of nominal exchange rate is popular for decades. A majority of papers focused on Monetary Model but failed to make a consistent conclusion. In this article, we suggest that the failure of monetary model might be coming from the basic assumption of taking different countries¡¦ assets as ¡§perfect substitutes¡¨. Under such circumstances, we introduce another model named as ¡§Portfolio Balance Model¡¨ where assets of different countries are no longer be taken as ¡§perfect substitutes¡¨ , implying that UIP( Uncoverd Interest Rate Parity)exist no more either. We do not overthrow the entire theory of Monetary Model. Instead, we expect the combination of these two models will turn something out that can be much more general, consistent, and robust.
We take Canada as our domestic currency and adopt Johansen(1988) and Stock & Walson(1988) by using co-integration to test on three exchange rates relation (USD/CAD,GBP/CAD,JPY/CAD) from 1973 Q1 to 2004 Q4. It turns out that most of the coefficient are correct and passing statistical significance, such result suggest that the portfolio balance effect should not be ignored in the model.
Advisor:Jia-hsi Wong; Yung-Hsiang Ying; Shan-non Chin
School:National Sun Yat-Sen University
School Location:China - Taiwan
Source Type:Master's Thesis
Keywords:cointegration portfolio balance model exchange rate determination
Date of Publication:06/26/2007