Empirical Properties of Foreign Exchange Rates
- PDF / 2,875,318 Bytes
- 11 Pages / 540 x 777 pts Page_size
- 38 Downloads / 225 Views
* Ina recent paper Westerfield[10] concludes that a symmetric stable distributionprovides INTRODUCTION a good description of exchange rate changes during fixed and floating rate regimes-certainly better than a normal distribution. The implications of this finding are extreme. For instance, if exchange rate changes follow a stable probabilitylaw, the sample variance is probably meaningless as a measure of risk. This implies that mean-variance models are of limited use in describing economic problems in internationalfinance. The stable model further implies that standard statistical tools-such as linear regression-may give misleading results which will have an impact on empirical tests of developed equilibriummodels. It is important to note that alternative distributionalhypotheses (with less extreme side-effects) besides the symmetric stable class were not examined by Westerfield. The aim of this paper is to extend the analysis of Westerfield by considering the Student t-distribution to explain exchange rate changes. There are good reasons for doing so. Recent work by Blattberg and Gonedes [2] and Praetz [8] has suggested that the Student distributionmay be superior to the symmetric stable distributionfor common stock returns. This makes it a logical candidate to consider for foreign exchange markets. There are other good reasons for investigating the distributionof exchange rate fluctuations. The interest rate paritytheory implies that there is a knowable 'normal" price in the exchange market. A divergence between observed prices and "normal" prices would imply inefficiency. Alternatively,if foreign exchange markets are efficient, futures prices "fully reflect" all available informationabout the future spot price. Investorsare assumed to act quickly on any informationthey receive so that exchange rates react quickly to the arrivalof information. Itis generally believed that foreign exchange markets are very efficient. Indeed, a well-functioning internationalmonetary system ought to be very efficient. Imperfectexchange markets would implyinefficiency. A common argument is that floating exchange rates destabilize exchange markets because speculators enter the marketto cre* RichardJ. Rogalskiis AssistantProfessorof Finance,the AmosTuckSchoolof BusinessAdministration,DartmouthCollege, wherehe is conductingresearchin capitaltheoryunderuncertainty. ** Joseph D. Vinsois AssistantProfessorof Finance,the WhartonSchool, Universityof Pennsylvania,whose researchinterestsinvolvecorporatefinancialmanagement. JaniceWesterfieldwas kindenough to lendherdatabase. Theauthorsgratefullyacknowledgethe assistanceof NicholasGonedesinobtainingthe computerprogramsnecessaryto completethispaper. Helpfulcommentsand suggestionswere providedby the anonymousreviewersof this journal. Researchsupportwas providedby the TuckAssociatesProgramandthe RodneyL.WhiteCenterfor FinancialResearch.
69
Palgrave Macmillan Journals is collaborating with JSTOR to digitize, preserve, and extend access to Journal of International Business Studies ® www.jstor.org
Data Loading...
