本文選題：購買力平價理論　切入點：實際匯率　出處：《北京外國語大學》2017年碩士論文　論文類型：學位論文

【摘要】：購買力平價理論是描述匯率變化的基本理論之一。前期的實證檢驗主要致力于探討購買力平價是否成立�；诰�性單位根檢驗對實際匯率進行分析,學者們發(fā)現(xiàn)很難拒絕數據存在單位根的原假設。但通過采用長時間跨度數據和面板數據在一定程度上可以拒絕存在單位根的原假設,進而證明購買力平價的有效性�？墒怯纱擞嬎愠龅膶嶋H匯率趨近長期均衡的調整速度非常緩慢,半衰期震蕩調整(half-life shock adjustment)大致需要三年到五年的時間。學者研究發(fā)現(xiàn)非線性模型能更有效的解釋購買力平價之謎,因此近些年基于TAR和STAR非線性模型的購買力平價實證研究一直是學術界熱議的話題。對于人民幣是否符合購買力平價理論,學者們也是莫衷一是。早期的線性單位根檢驗和協(xié)整檢驗無法證實購買力平價理論的成立,甚至有學者認為我國目前的經濟狀況并不適用PPP理論。對此,許多學者試圖探尋人民幣匯率偏離PPP的原因,也有的學者通過構建非線性框架和模型來分析人民幣匯率的波動。本文基于非線性均值回復模型ESTAR分析了英鎊-美元、馬克-美元、法郎-美元和人民幣-美元的實際匯率月度數據,探討并對比了其非線性均值回復的特征。通過研究人民幣-美元實際匯率的變動進一步探討了購買力平價在我國當前經濟環(huán)境中的適用性,同時依據人民幣實際匯率的實證檢驗結果提出了相應的政策建議。本文有以下創(chuàng)新點。首先,因為現(xiàn)有文獻較少考慮到異方差,并基于殘差的特征,本文將ESTAR模型進一步拓展為ESTAR-GARCH,同時還引入了虛擬變量來刻畫匯率制度變更對實際匯率的影響。其次,本文補充了人民幣匯率的相關的實證研究。因為基于人民幣匯率的實證檢驗較少支持購買力平價理論,且基于ESTAR模型分析人民幣匯率非線性特征的研究較少,因此希望可以通過本文補充相關的實證研究。再次,在進行樣本外循環(huán)預測時,本文將英鎊-美元、馬克-美元、法郎-美元實際匯率的樣本外預測期間擴展為整個后布雷頓森林體系時期,人民幣-美元的實際匯率的樣本外預測期間為2005年7月匯率制度改革至今。本文基于生成的預測值兩兩對比了隨機游走模型、AR模型、ESTAR和ESTAR-GARCH模型對于實際匯率變動的預測能力。本文分析了英鎊-美元、馬克-美元、法郎-美元和人民幣-美元實際匯率后,發(fā)現(xiàn)實際匯率的時間序列存在非線性�；貧w結果證實了用ESTAR和拓展后的ESTAR-GARCH模型來刻畫實際匯率非線性動態(tài)過程的有效性。本文還通過模型預測精度檢驗(forecast accuracy test)和(forecast encompassing test)預測包容檢驗,兩兩對比了隨機游走模型、線性AR模型、非線性ESTAR模型和ESTAR-GARCH模型的預測效果。檢驗結果表明非線性ESTAR和ESTAR-GARCH模型對于實際匯率變動的預測能力都要優(yōu)于隨機游走模型。具體就人民幣匯率而言,本文基于從1987年至2016年的人民幣-美元實際匯率實證檢驗結果表明購買力平價理論適用于當前的人民幣實際匯率,且實際匯率呈現(xiàn)出非線性均值回復的特征。該特征的產生主要是由于我國對匯率變動的宏觀干預和調整所致。雖然回歸結果可以看出在四組時間序列數據中,人民幣呈現(xiàn)的非線均值回復特征最典型,且實際匯率在受到沖擊后趨向長期均衡的速度最快,反映出我國匯率制度改革成效顯著。但是因為我國資本市場開放程度較低,且人民幣國際化是我國中長期發(fā)展目標,匯率制度改革仍任重道遠。因此要繼續(xù)增強人民幣匯率的彈性,逐步實現(xiàn)人民幣匯率的自由浮動。同時,從短期人民幣-美元匯率可以看出,人民幣的確存在一定程度的高估,長期還是需要進一步深化匯率機制改革,推進人民幣資本項目可兌換。最后還需要建立全面的宏觀經濟監(jiān)測分析框架,以免發(fā)生恐慌性的資本外逃。通過穩(wěn)定人民幣的匯率,減少外部沖擊帶來的影響,進而推動人民幣的國際化。
[Abstract]:The PPP theory is one of the basic theory to describe the changes in the exchange rate. The empirical test focused on the PPP. The linear unit root test to analyze the real exchange rate based on the scholars found it difficult to reject the null hypothesis of unit root data. But by using long time span and panel data in a certain the original extent can reject the unit root hypothesis, and then prove the validity of the PPP. But the real exchange rate to calculate the adjustment speed reaching long-term equilibrium is very slow, the half-life of the shock adjustment (half-life shock adjustment) takes roughly three years to five years. The researchers find that the nonlinear model can effectively explain the PPP puzzle, so in recent years, the empirical research on the TAR and STAR nonlinear model based on purchasing power parity has been the academic hot words For questions. Whether the RMB accords with the theory of purchasing power parity, scholars are unable to agree on which is right. Linear unit root test and cointegration test to early established theory of purchasing power parity, even some scholars believe that China's current economic situation does not apply PPP theory. In this regard, many scholars have tried to explore the reasons for the RMB exchange rate deviates from the PPP, also some scholars by constructing nonlinear framework and model to analyze the volatility of the RMB exchange rate. This paper analyzes the nonlinear mean reversion model ESTAR Sterling dollar based on Mark dollar real exchange rate, the monthly data of the dollar and the renminbi dollar - Franc, discussed and compared the characteristics of nonlinear mean reversion. Through the study of the real exchange rate of RMB to US dollars changes to further explore the applicability of PPP in China's current economic environment, at the same time on the basis of empirical real exchange rate of RMB The test results and puts forward some policy suggestions. This thesis has the following innovative points. First, because the existing literature less considering heteroskedasticity, and based on the residual characteristics, this paper will further expand the ESTAR model for ESTAR-GARCH, but also the introduction of dummy variables to describe the exchange rate change impact on the real exchange rate. Secondly, this paper adds Empirical Study on the RMB exchange rate. Because the empirical test of RMB exchange rate with less support based on PPP theory, and based on ESTAR model to analyze the nonlinear characteristics of RMB exchange rate is less, therefore the hope can supplement the relevant empirical study. Through this again, out of sample forecasts in the cycle, the sterling dollar, Mark dollars, francs - dollar real exchange rate for the entire sample forecast during the expansion of the post Bretton Woods period, the real exchange rate of RMB - dollar Out of sample forecast period is since the reform of exchange rate system in July 2005. The forecast based on the value of 22 compared to the random walk model, AR model, ESTAR model and ESTAR-GARCH model for predicting the ability of real exchange rate changes. This paper analyzes the sterling dollar, Mark - dollars, francs - dollar and Renminbi dollar real exchange rate after the discovery the time series of the real exchange rate is nonlinear. The regression results confirmed the effectiveness of ESTAR-GARCH model with ESTAR and expanded to describe the nonlinear dynamic process of the real exchange rate. This paper also through the accuracy test prediction model (forecast accuracy test) and (forecast encompassing test) 22 compared the forecast encompassing test, random walk model, linear AR model. The prediction effect of nonlinear ESTAR model and ESTAR-GARCH model. The test results show that the nonlinear ESTAR and ESTAR-GARCH models for the real exchange rate change The forecasting ability is better than the random walk model. The RMB exchange rate, this paper from 1987 to 2016 the RMB real exchange rate dollar - based on the empirical results show that PPP theory applies to the current real exchange rate of RMB real exchange rate, and showing the characteristics of nonlinear mean reversion. This feature is caused by me China due to exchange rate changes and the adjustment of macro intervention. Although the regression results can be seen in the four groups of time series data, the present RMB non line mean reverting the most typical characteristics, and the real exchange rate in the impact trend of long-term equilibrium is the fastest, reflecting the effectiveness of China's exchange rate system reform is significant. But because of the degree of openness China's capital market is relatively low, and the RMB internationalization is our long-term development goals, the reform of exchange rate system still has a long way to go. So it will continue to increase Strong RMB exchange rate flexibility, the progressive realization of the RMB exchange rate to float freely. At the same time, from the short-term RMB dollar exchange rate of RMB is overvalued can be seen, to some extent, the long-term or the need to further deepen the reform of exchange rate mechanism, and promote capital account convertibility. Finally, also need to establish a comprehensive framework to analyze the macroeconomic monitoring, in order to avoid panic capital flight. The stability of RMB exchange rate, reduce the impact of external shocks, thus promoting the internationalization of the RMB.

【學位授予單位】：北京外國語大學
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：F832.6

【參考文獻】

相關期刊論文 前10條

1 宋保慶;;基于STAR模型的人民幣實際有效匯率波動分析[J];金融發(fā)展研究;2016年09期

2 王正新;陳雁南;;“匯改”以來人民幣實際有效匯率波動的非線性機制——基于平滑轉移自回歸模型的實證研究[J];金融理論與實踐;2016年06期

3 戴金平;楊珂;劉東坡;;人民幣匯率對購買力平價的偏離及原因分析[J];中央財經大學學報;2015年07期

4 危黎黎;李余輝;李超;;人民幣匯率的非線性特征研究[J];中南財經政法大學學報;2014年02期

5 王磊;范超;;購買力平價與匯率背離原因研究[J];數量經濟技術經濟研究;2013年11期

6 趙留彥;黃桂田;;交易成本與長期購買力平價:近代中國的經驗證據[J];金融研究;2013年08期

7 王娟;;購買力平價及在亞洲新興市場的實證分析[J];統(tǒng)計研究;2012年03期

8 楊長江;鐘寧樺;;購買力平價與人民幣均衡匯率[J];金融研究;2012年01期

9 王熙;趙曉波;;人民幣匯率高頻數據非線性模型表現(xiàn):預測比較[J];商業(yè)經濟;2010年03期

10 趙振全;劉柏;;中國購買力平價的STAR非線性單位根檢驗[J];寧波大學學報(人文科學版);2008年06期

，

本文編號：1591095