本論文研究外匯市場中三項重要的議題：價格發現、匯率的波動、與中央銀行干預效果之研究。本論文以門檻共整模型分析新台幣外匯市場中，誤差修正過程的不對稱調整特性；並以具有跳躍(jump)型式的GARCH模型分析歐元匯率在不同流通階段波動性的變化，以及日本央行藉由外匯市場干預行動對日圓匯價與波動性造成的影響效果。本論文重要發現如下： 1、新台幣匯率的調整具有非線性特性。台北外匯市場與元太外匯市場的價格具有長期共整關係，然此關係具有門檻效果：唯有當兩市場價差超過上、下門檻值時，兩市場價格才具有互動關係。並且當元太市場價格大於台北市場價格，且價差已超越門檻值後，元太市場價格的弱外生性顯示元太市場具有價格發現功能，台北市場將追隨元太市場的價格動態而進行誤差調整。 2、 歐元匯率的波動性在本文研究期間大於比較樣本—英鎊的波動性，顯示歐元資產的投資風險仍大於英鎊資產。然分期間研究結果顯示，兩者波動性的差距已有縮小跡象。代表當歐元流通區域擴大及與其他經濟體關係日益密切等因素下，已有助於實體歐元流通之後的穩定性。 3、中央銀行干預的型態會影響干預效果。小量干預不具成效，大量干預可影響匯率走勢並降低市場波動。在大量干預時期，日銀成功透過干預行動降低市場跳動次數，故連帶降低市場因日圓升值帶來的擾動現象。此結果符合Taylor (2004)、Reitz & Taylor (2006)提出之coordination channel：央行干預行動可提振基本面交易者的交易信心，使匯率重回基本面，而市場重歸平靜。 Three topics in exchange rate economics are discussed in this thesis: the price discovery in foreign exchange markets, volatility and the effects of central bank interventions. A threshold cointegration model proposed by Tsay (1998) is used to analysis the asymmetric adjustment process responding to the disequilibrium between two market places in Taiwan’s foreign exchange trading. A GARCH mixed with jump model proposed by Chan and Maheu (2002) is applied to detect the volatility changes of a new currency – the euro, and to evaluate the effectiveness of foreign exchange market intervention conducting by the Japanese authorities. Major conclusions are as follows. 1.Results from the multivariate threshold model indicate prices in the two markets in Taiwan’s foreign exchange trading are integrated nonlinearly. The roles of price discovery are asymmetric, depending on the size and sign of the price discrepancies between the two markets. In the lower regime of discrepancies, each market employs information from its counterpart and reacts to each other with different adjustment speeds. When the discrepancy is in the upper regime, Cosmos Foreign Exchange’s role of price discovery is characterized by its exogenous behavior within the error-correction process. 2.By using a GARCH mixed with jump model (the GARJI model), the volatilities of euro are found to be larger than that of the pound, which is robust to either of the two data-splitting schemes. However, the stability of the euro has made progress in recent years when the physical euro launched in circulating. The evidence is provided by the decreases in the jump innovations. This finding supports the arguments on the determinants of exchange rate stability claimed by Mundell (1998) and Mussa (2000). 3.The empirical results show ‘large-in-size’ interventions adopted by Bank of Japan in the last decade were effective both in altering the exchange rate level and reducing the volatility of Yen. Jump events that tended to drive yen’s appreciations and volatility increases have been effectively reduced, which are in compliance with the coordination channel that explains the effectiveness in intervention proposed by Taylor (2004) and Reitz and Taylor (2006).