應用外匯市場微結構理論建構美元兌新台幣價差、價量之關係

Applying the Micro-Structure Theory to Construct the Interrelationship Among Bid-Ask Spread, Return, and Trading Volume in the NTD/USD Currency Marks Markets

王素雲(WANG SU-YUN)

指導教授 ： 陳達新 ； 林婷鈴

國立臺北大學/商學院/企業管理學系/碩士(2011年)

摘要

隨著國際化潮流及網路資訊的迅速發展，國際貿易及投資愈加頻繁。台灣屬海島型經濟國家體系，對外貿易依存度甚高，且近年來政府大幅放寬外匯管制及大力推動國際化與自由化，使得新台幣匯率波動更加劇烈，但一國之匯率若過度波動，容易造成該國經濟的動盪不安。
長期以來，經濟學家皆將種種影響匯率的波動原因，歸諸於總體經濟現象。但是近年來，由於國際資本的高度流動，加上貨幣風暴頻傳，使得以往運用傳統總體經濟解釋匯率的變因已明顯不足。2008年世紀金融海嘯事件的發生，世界多數國家面臨經濟危機，各國政府為求經濟穩定成長，競相以貨幣貶值作為工具，造成外匯市場匯率偏離正常運作的軌道，此種種匯率波動現象，皆非傳統總體經濟學所能解釋。因此，已有部分學者開始將運用於證券投資成效良好的微結構理論引入外匯市場，Loyns (2001) 發表的“The Microstructure Approach to Exchange Rates”，顯示微結構理論的基本框架已經形成，並陸續有眾多的文獻探討匯率微結構的成因，且不斷發現傳統總體經濟無法解釋的謎團，包括外匯市場私有信息、交易者的異質性及交易機制等微結構領域。
本研究應用外匯市場微結構理論探討價差、價量之關係，樣本範圍跨越金融海嘯事件期間，故將樣本區分為金融海嘯事件前、金融海嘯事件後、及全部樣本期間。並以GARCH-GED模型建構價差及匯率報酬率之關係，運用二元GARCH模型探討價量間之互動，結果發現價差及匯率報酬率間具有相關性，匯率及成交量亦有領先、落後之關係，其中金融海嘯事件並未改變前期匯率變動率及交易量變動率對當期匯率變動率之影響方向，分別為負相關及無相關；前期匯率變動率及交易量變動率對當期交易量變動率亦未受金融海嘯事件影響而改變，兩者間呈負相關。

關鍵字

匯率 ； 市場微結構理論 ； 金融海嘯事件 ； 價差 ； 價量關係

並列摘要

Globalization and rapid advances in online information have accelerated the tempo of international trade and investment activity. As an island economy, Taiwan is highly dependent on foreign trade. In addition, the government in recent years has greatly eased foreign exchange restrictions and vigorously pursued globalization and liberalization, which has increased the volatility of the New Taiwan Dollar exchange rate. Excessive volatility in a country’s exchange rate can easily result in economic instability.
For quite some time now, economists have generally attributed exchange rate fluctuations to macroeconomic phenomena, but in recent years, big movements of international capital and frequent currency crises have made it very clear that the traditional macroeconomic approach is an inadequate means of explaining the causes of exchange rate movements. The financial tsunami of 2008 triggered economic crises in most countries throughout the world, and governments used currency devaluation as a tool to seek stable economic growth. This knocked foreign exchange rates off their normal course, and the exchange rate fluctuations seen since then cannot be explained through traditional macroeconomics. Microstructure theory, which was previously applied very successfully to securities investing, has therefore been adapted by some scholars to the study of foreign exchange markets. The publication of The Microstructure Approach to Exchange Rates (Lyons, 2001) signaled the formation of a basic framework for microstructure theory, and many academic works published since then have further explored exchange rate microstructure. In the process, scholars have focused on many matters that cannot be explained by traditional macroeconomics, including proprietary information in foreign exchange markets, heterogeneity among traders and trading mechanisms, and other microstructural issues.
This study applies foreign exchange market microstructure theory to a discussion of price spreads and price-volume relationships. The sample set includes the period of the financial tsunami, and so has been broken down into a pre-tsunami subset, post-tsunami subset, and a single complete set. The GARCH-GED model is used to establish a relationship between price spreads and exchange rate returns, and a bivariate GARCH model is used to discuss price-volume interaction. The resulting discovery is a correlation between price spreads and exchange rate returns, as well as a leading and lag relationship between exchange rates and trading volumes. The financial tsunami did not change the directional impact of the previous period exchange rate volatility and transaction volume fluctuation rate upon the current period exchange rate volatility, which is inversely correlated in the former case and uncorrelated in the latter. In addition, the impact of the previous period exchange rate volatility and transaction volume fluctuation rate upon the current period transaction volume fluctuation rate was not affected by the financial tsunami, with a negative correlation between both pairs.

並列關鍵字

exchange rates, market microstructure theory, financial tsunami, price spreads, price-volume relationship

參考文獻

高櫻芬(2006)，外匯市場的買價與賣價的動態調整過程：以日圓/美元市場為例，中山管理評論，頁873-896。

張元晨(2007)，銀行間新臺幣兌美元外匯交易流動性與交易成本的分析：臺北與元太外匯經紀公司的比較，中山管理評論，頁299-321。

高櫻芬、彭雅玲(2003)，外匯市場之市場微結構，台灣金融財務季刊，頁103-115。

陳達新、周恆志(2010)，財務風險管理，雙葉書廊。

Akaike, H. (1976). Canonical correlation’s analysis of time series and the use of an informationa criterion. The Institute of Statistical Mathematics, 126, 27-97.