請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/79070
標題: | 在岸與離岸人民幣匯率差異影響因素之研究——基於時變結構與ARDL模型的實證 A Study of the Determinants of CNH-CNY Pricing Differentials:Evidence from Time-Varying Structures and ARDL Model |
作者: | Rong Yang 楊融 |
指導教授: | 黃志典(Jyh-Dean Hwang) |
關鍵字: | 在岸與離岸人民幣匯率差異,影響因素,時變結構,BREAKLS模型,ARDL模型, CNH-CNY pricing differentials,Determinants,Time-varying structures,BREAKLS model,ARDL model, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 隨著人民幣國際化的推進,離岸人民幣市場的建設成為重要一環。然而,自2011年香港财资市场公会推出美元兑人民币(香港)即期汇率以来,在岸與離岸人民幣市場的匯率差異就長期顯著存在,带来了许多套利机会。
長期看來,根據一價定律,同種商品在不同地區的價差會由於套利活動消失。價差的長期存在是由於價格粘性、交易成本與資本流動障礙。本文結合現行具體政策與市場環境深入探討,發現在在岸與離岸人民幣市場化程度的不同與資本管制是匯差長期存在的根本原因。對於相關政策,本文運用虛擬變量建立OLS回歸模型,考量具體政策在縮小匯差上所發揮的效果。 影響匯差短期變化因素的探討同樣重要。本文基於匯率決定理論,選取在岸與離岸人民幣利差、離岸人民幣與美元利差的同時,引入升貶值預期、美元指數、恐慌指數這五個變量。由於政策與市場環境的變化較大,匯差影響因素的作用機制也非恆定,因此本文運用BREAKLS模型與ARDL模型,分階段探討各變量對匯差的影響。研究發現人民幣升貶值預期與前一日匯差是影響當日匯差的重要原因,而從各階段變量的總體影響來看,中美利差與人民幣升貶值預期的影響方向較為穩定。 基於ARDL模型的結果,本文引入衝擊反應函數和半衰期的概念,分析各階段在岸與離岸人民幣市場的一體化程度。分析得到匯差在各階段都具有自動收斂機制,但收斂速度各不相同。目前在岸與離岸人民幣市場一體化程度已經較高。 根據上述分析,本文認為,對於投資者來說套利的獲利空間已大大減小,但仍然存在,而人民幣升貶值預期與前一日匯差可以作為較好的參考指標;對於中國人民銀行來說,短期可以通過穩定預期和調整利率來穩定匯差,長期則需繼續推進在岸人民幣市場的市場化改革和資本自由流動,根本上穩定匯差在合理範圍內。 With the rapid development of yuan-internationalization, Hong Kong’s offshore market has become more and more important. Nevertheless, the differential between the CNH and CNY exchange rates has been overwhelmingly distinct since the USD/CNY(HK) Spot Rate was launched by Hong Kong Treasury Markets Association in 2011, which brings out numerous arbitrage opportunities. In the long-term, the differential will be reduced by arbitrage according to the law of one price. The existence of the pricing differential is due to price stickiness, transaction costs and barriers to the free flow of capital. We find that the differences in the degree of marketization between the two markets and capital controls are the fundamental cause of the differential. The concrete impacts of the factors are measured using the OLS model. Meanwhile, the study of the short-term causes of the differential is also important. The lagged variables of the CNH-CNY interest rate spreads, the CNH-USD interest rate spreads, the expectation of RMB, US Dollar Index and CBOE Volatility Index are introduced as exogenous variables. The lagged variables of the pricing differentials are also included as endogenous variable. The BREAKLS and ARDL model are used because of the unknown breakpoints and lags. It is found that the differential in one day is mainly decided by the differential of one day before and expectation of the day. The overall positive impact of the expectation of RMB is significant. Based on the ARDL model, half-life and impulse response function are used to measure the onshore-offshore RMB market integration in different courses. The auto-convergence mechanism of the pricing differential is significant but the convergence speed differs within different phases. The conclusion is that the degree of integration between onshore and offshore markets is high at this stage. Therefore, the arbitrage space is largely squeezed but still exists. The investors can refer to the expectation and the differential of one day before to arbitrage. For People’s Bank of China, stabilizing expectations and interest rate adjustment can help to keep the differential in the certain range. Most fundamentally, the main tasks for PBC to reduce the differential are to further the onshore market reforms and gradually loosen the capital control. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/79070 |
DOI: | 10.6342/NTU201803105 |
全文授權: | 有償授權 |
電子全文公開日期: | 2023-08-21 |
顯示於系所單位: | 國際企業學系 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-R05724068-1.pdf 目前未授權公開取用 | 1.01 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。