Topic 6: Exchange Rate Predictability
主题 6：汇率可预测性

Factors affecting the exchange rate between US Dollar and Chinese RMB
影响美元兑人民币汇率的因素

Table of Contents
目录

Chapter 1 Introduction 2
第 1 章 引言 2

1.1. Background 2
1.1. 背景 2

1.2. Research Questions and Objectives 3
1.2. 研究问题和目标 3

1.3. Research rationales 4
1.3. 研究理由 4

Chapter 2 Literature Review 5
第 2 章 文献综述 5

2.1. Concept of Exchange Rates 5
2.1. 汇率的概念 5

2.2. Theoretical Foundations 5
2.2. 理论基础 5

2.3. Empirical Evidence 6
2.3. 经验证据 6

Chapter 3 Research Methods 8
第 3 章 研究方法 8

3.1. Research Design 8
3.1. 研究设计 8

3.2. Data Collection 8
3.2. 数据收集 8

3.3. Empirical Model 9
3.3. 经验模型 9

Chapter 4 Findings and Evaluation 11
第 4 章 调查结果和评估 11

4.1. Descriptive Analysis 11
4.1. 描述性分析 11

4.2. Correlation Analysis 13
4.2. 相关性分析 13

4.3. Regression Analysis 15
4.3. 回归分析 15

4.4. Evaluation of Findings 16
4.4. 调查结果评估 16

Chapter 5 Conclusion 17
第 5 章 结论 17

5.1. Summaries of Findings 17
5.1. 调查结果摘要 17

5.2. Research Limitations 18
5.2. 研究局限性 18

5.3. Suggestions for Future Research 19
5.3. 对未来研究的建议 19

References 21
参考资料 21

Chapter 1 Introduction
第 1 章 引言

Background
背景

Exchange rates is an important part of the global economy, influencing international trade flows, investment decisions, inflation dynamics, and monetary policy effectiveness. Exchange rates are fundamental to the functioning of the international financial system. They determine the relative value of one currency against another, influencing the cost of imports and exports, the attractiveness of investment opportunities, and the competitiveness of economies. A country's exchange rate regime whether floating, fixed, or under management can shape its monetary sovereignty and its exposure to global financial volatility. For businesses, the changes in exchange rates can have an impact on profit margins, strategies related to pricing, and supply chain planning. For governments, they influence inflation, interest rates, and fiscal stability. Moreover, exchange rates have a vital role in capital mobility and investment flows, as currency expectations influence decisions on foreign direct investment as well as cross-border lending. A strong or appreciating currency can signal economic strength and attract investment, while excessive volatility or depreciation may deter investors and signal macroeconomic risk. In emerging and developing economies, such as China, exchange rate movements are also tied to exchange rate pass-through effects, which impact domestic inflation and household purchasing power.
汇率是全球经济的重要组成部分，影响着国际贸易流动、投资决策、通货膨胀动态和货币政策有效性。 汇率是国际金融体系运作的基础。它们决定了一种货币相对于另一种货币的相对价值，影响进出口成本、投资机会的吸引力和经济竞争力。一个国家的汇率制度是浮动的、固定的或管理下的 ， 可以塑造其货币主权及其对全球金融政策的敞口 。对于企业来说，汇率的变化会影响利润率、 与定价相关的策略和供应链规划。对于政府来说，它们影响通货膨胀、利率和财政稳定。 此外，汇率 在资本流动性和投资流动中起着至关重要的作用，因为货币预期会影响外国直接投资和跨境贷款的决策。强势或升值的货币可以表明经济实力并吸引投资，而过度波动或贬值可能会阻止投资者并预示着宏观经济风险。在中国等新兴和发展中经济体，汇率变动也与汇率转嫁效应有关，从而影响国内通胀和家庭购买力。

Among these, the Renminbi (RMB)–US Dollar (USD) exchange rate has garnered substantial scholarly and policy interest, given China's growing economic influence and the dominance of the USD as the world’s reserve currency. The USD continues to be the most widely used reserve currency in the world, supporting a large portion of global financial and commercial exchanges. Because it is used globally, the US dollar is thought to be a more stable currency. The greatest increase in inflation in more than 40 years was caused by the US adopting low interest rates and limitless quantitative easing, tightening the banking system, and slowing economic development. These events, along with the effects of the pandemic and the US financial crisis, caused the US dollar exchange rate to fluctuate significantly. In addition, China has the second-biggest economy in the world and is becoming more interdependent. The global economy has been significantly impacted by the volatility of the RMB exchange rate as China's overseas engagement continues to deepen. The RMB has increased at a startling rate in recent years. The US dollar to RMB exchange rate is nearly at its peak till 2021. This pattern is partly a result of China's robust economic expansion and the RMB's acceptance on the global market. At the same time, China's economic rise, growing share of global trade, and increasing internationalization of the RMB have led to a surge in scholarly and policy interest in understanding and predicting the behavior of the RMB–USD exchange rate. The People's Bank of China's gradual shift from a fixed exchange rate regime to a more flexible, market-oriented mechanism, albeit still managed, has added further complexity and significance to exchange rate forecasting. Furthermore, it is still difficult to consistently anticipate exchange rate movements despite decades of research, and there is ongoing debate on the macroeconomic factors that affect exchange rate volatility and trends. Purchasing Power Parity (PPP), Uncovered Interest Parity (UIP), and monetary models are examples of foundational models based on macroeconomic fundamentals that have frequently fallen short of naïve standards like the random walk model, particularly in out-of-sample projections. The degree and circumstances of exchange rate predictability are still up for dispute in the field of international finance as a result of this discrepancy.
其中， 鉴于中国日益增长的经济影响力以及美元作为世界储备货币的主导地位，人民币 （RMB） 兑美元 （USD） 汇率引起了学术界和政策界的广泛兴趣。 美元仍然是世界上使用最广泛的储备货币，支持着大部分全球金融和商业交易所。因为它在全球范围内使用 ，所以美元被认为是一种更稳定的货币。40 多年来最大的通货膨胀增幅是由于美国采取低利率和无限量化宽松、收紧银行体系和减缓经济发展造成的 。 这些事件，加上大流行和美国金融危机的影响，导致美元汇率大幅波动。 此外，中国是世界第二大经济体， 并且相互依存性越来越强。 随着中国海外业务的持续深入， 全球经济受到人民币汇率波动的重大影响 。近年来， 人民币以惊人的速度上涨。直到 2021 年 ，美元兑人民币的汇率几乎处于峰值 。这种模式部分是由于中国强劲的经济扩张和人民币在全球市场上的接受度。 与此同时，中国经济的崛起、全球贸易份额的扩大以及人民币的日益国际化 ，导致学术和政策界对理解和预测人民币 -美元汇率行为的兴趣激增。中国人民银行逐渐从固定汇率制度转变为更加灵活 、市场化的机制，尽管仍然受到管理， 这进一步增加了汇率预测的复杂性和重要性 。 此外， 尽管进行了数十年的研究， 但仍然难以始终如一地预测汇率变动，并且关于影响汇率波动和趋势的宏观经济因素的争论仍在继续。购买力平价 （PPP）、未覆盖利息平价 （UIP） 和货币模型是基于宏观经济基本面的基础模型的示例，这些模型经常达不到随机游走模型等幼稚标准，尤其是在样本外预测中。由于这种差异，汇率可预测性的程度和情况在国际金融领域仍存在争议 。

Seminal work by Meese and Rogoff (1983) challenged the effectiveness of structural macroeconomic models by demonstrating that they failed to outperform a naïve random walk in out-of-sample forecasting. Subsequent studies attempted to address these limitations by refining the models, introducing real-time data, and adopting non-linear and time-varying methodologies. For instance, Cheung, Chinn, and Pascual (2005) conducted a comprehensive comparison of traditional and modern models across multiple currencies and found that while some models provided marginal improvements over the random walk, predictive gains were often limited to specific currencies or time horizons. Similarly, Rossi (2013) highlighted the importance of model stability and robustness checks, arguing that while some models achieve better short-run predictions during periods of market stress, their performance can be inconsistent over longer horizons. More recent empirical work has focused on macro-financial linkages and the role of global shocks. For example, Engel and West (2005) proposed that exchange rates may be forward-looking asset prices that reflect expectations of future fundamentals, explaining their apparent disconnect from current economic indicators. Meanwhile, Kilian and Taylor (2003) introduced nonlinear models to capture regime shifts in exchange rate behavior, particularly during crises and periods of central bank intervention.
Meese 和 Rogoff （1983） 的开创性工作通过证明结构宏观经济模型在样本外预测中未能胜过天真的随机游走，挑战了结构宏观经济模型的有效性。随后的研究试图通过改进模型、引入实时数据以及采用非线性和时变方法来解决这些限制。例如，Cheung、Chinn 和 Pascual （2005） 对多种货币的传统和现代模型进行了全面比较，发现虽然一些模型在随机游走中提供了边际改进，但预测收益通常仅限于特定货币或时间范围。同样，Rossi （2013） 强调了模型稳定性和稳健性检查的重要性，认为虽然一些模型在市场压力期间实现了更好的短期预测，但它们的性能在长期内可能不一致。最近的实证工作主要集中在宏观金融联系和全球冲击的作用上。例如，Engel 和 West （2005） 提出汇率可能是反映未来基本面预期的前瞻性资产价格，解释了它们与当前经济指标的明显脱节。与此同时，Kilian 和 Taylor （2003） 引入了非线性模型来捕捉汇率行为的制度变化，尤其是在危机和央行干预期间。

Overall, while the consensus remains that exchange rate forecasting is inherently difficult, especially in the short term, empirical advancements continue to refine our understanding. The combination of high-frequency data, machine learning techniques, and dynamic factor models is now at the forefront of research, opening new pathways for more effective forecasting.
总体而言，虽然人们仍然认为汇率预测本质上是困难的，尤其是在短期内， 但实证进步继续完善我们的理解。高频数据、机器学习技术和动态因子模型的结合现在处于研究的前沿，为更有效的预测开辟了新的途径。

Research Questions and Objectives
研究问题和目标

A research is proposed to answer the following question: “To what extent are exchange rates between the US dollar and Chinese yuan predictable based on macroeconomic fundamentals including inflation, interest rate and current account balance under China’s managed floating system?”
本研究提出一项研究 来回答以下问题：“ 在 China 的管理浮动系统下， 美元和人民币之间的汇率在多大程度上可以预测宏观经济基本面，包括通货膨胀、利率和经常账户余额 ？

The objectives of the research are
研究目标是

To identify and empirically evaluate the key macroeconomic, financial, and geopolitical variables that influence RMB–USD exchange rate volatility
识别和实证评估影响 RM B-USD 汇率波动的关键宏观经济、金融和地缘政治变量

To assess the in-sample and out-of-sample predictive power of various exchange rate models
评估各种汇率模型的样本内和样本外预测能力

To generate policy-relevant insights for central banks and international investors, particularly regarding the exchange rate forecasting
为中央银行和国际投资者提供与政策相关的见解，特别是关于汇率预测的见解

Research rationales
研究理由

International finance researchers have debated extensively about the exchange rate predictability. A multitude of years of research has failed to establish whether macroeconomic models achieve better predictive power than simple random walk models. The analysis concentrates on the US-Chinese economies because of their economic powers and conflicts in recent years. It would be interesting to assess the level of exchange rate policies' sensitivity while testing if macroeconomic indicators including interest rates inflation and current account balances can effectively predict currency rate fluctuations.
国际金融研究人员对汇率的可预测性进行了广泛的辩论。多年的研究未能确定宏观经济模型是否比简单的随机游走模型具有更好的预测能力。该分析集中在美中经济上，因为它们的经济实力和近年来的冲突。在测试包括利率、通货膨胀和经常账户余额在内的宏观经济指标是否能够有效预测货币汇率波动的同时，评估汇率政策的敏感性水平会很有趣。

The project will make several contributions to the field of international finance. First, it will offer updated empirical evidence on the predictability of exchange rates using recent data. New data is expected to reflect changes of the economic upheavals caused by the COVID-19 pandemic and US-China trade tensions. Second, the study will provide a comparative evaluation of different forecasting models, offering insights into their relative effectiveness and limitations.
该项目将为国际金融领域做出多项贡献 。首先，它将使用最近的数据为交货率的可预测性提供最新的实证证据 。预计新数据将反映 COVID-19 大流行和美中贸易紧张局势造成的经济动荡的变化。其次，该研究将对不同的预测模型进行比较评估，提供对其相对有效性和局限性的见解。

Chapter 2 Literature Review
第 2 章 文献综述

2.1. Concept of Exchange Rates
2.1. 交易所的概念

The link between supply and demand should determine the equilibrium price of an exchange rate, which is the rate of exchange between two currencies. However, when there is a rise in "speculative demand" for a currency as a result of arbitrage in predicted exchange rates, exchange rates may become skewed. This can lead to exchange rate bubbles, a phenomena when a currency's price significantly deviates from its true worth as a result of a steady rise in speculative capital (Hu and Oxley, 2017). According to Bettendorf and Chen (2013), an exchange rate bubble occurs when the value of the currency surpasses its intrinsic value, which is another significant aspect that contributes to the formation of economic bubbles. Sometimes, after a prolonged period of deviation from their long-term equilibrium, exchange rates abruptly self-correct. There are macroeconomic ramifications to these "unjustified" exchange rate fluctuations (Alexandre et al., 2011).
供求之间的联系应该决定汇率的均衡价格，即两种货币之间的汇率。但是，当由于预测汇率的套利而导致对某种货币的“投机需求”增加 时，汇率可能会变得偏斜。这可能导致汇率泡沫， 当货币价格由于投机资本的稳步增长而显着偏离其真实价值时（胡 和 Oxley，2017 年）。 根据 Bettendorf 和 Chen （2013） 的说法，当货币的价值超过其内在价值时，就会出现汇率泡沫，这是导致经济泡沫形成的另一个重要方面。有时，在长时间偏离其长期均衡后，汇率会突然自我纠正。这些“不合理”的汇率波动会产生宏观经济影响 （Alexandre et al.， 2011）。

Since 2009, changes in the Chinese economy and US financial and economic policies have had a significant impact on the volatility of currency rates in emerging economies (Liu et al., 2019). In worldwide banking and trade, the US dollar is vital. Since US dollars are used to settle a large number of international commerce and financial transactions, they serve as the main medium via which global currency moves.
自 2009 年以来，中国经济的变化以及美国的金融和经济政策对新兴经济体货币汇率的波动性产生了重大影响（Liu et al.， 2019）。在全球银行和贸易中，美元至关重要。由于美元 用于结算大量国际商业和金融交易，因此它们是全球货币流动的主要媒介。

2.2. Theoretical Foundations
2.2. 理论 Foundations

The literature on exchange rate determination and forecasting is extensive but remains inconclusive. Early foundational theories such as Purchasing Power Parity (PPP), Uncovered Interest Parity (UIP), and the monetary models of exchange rate determination (Frenkel and Mussa, 1976) proposed a strong linkage between macroeconomic fundamentals and exchange rate behaviour.
关于汇率确定和预测的文献广泛，但仍然没有定论。早期的基础理论，如购买力平价 （PPP）、未覆盖的利率平价 （UIP） 和汇率决定的货币模型（Frenkel 和 Mussa，1976 年）提出了宏观经济基本面与汇率行为之间的密切联系 。

Purchasing Power Parity (PPP), one of the most well-known ideas, holds that exchange rates should be adjusted to bring two nations' identical baskets of commodities to the same price. According to PPP in its absolute version, the ratio of domestic to international price levels should be the same as the nominal exchange rate. According to the relative form of PPP, which is more frequently applied in empirical research, the difference in inflation rates between two nations is equivalent to the pace at which a currency depreciates or appreciates. The fundamental premise is that cross-border price convergence is guaranteed by arbitrage through trade and money movements. However, PPP often holds up better over time, with short-term variations brought on by non-tradables, market frictions, and transaction costs. Froot and Rogoff (1995) argue that deviations from PPP can persist for years, especially in developing or transitioning economies. In the case of China, strict controls over pricing and capital flows further dampen PPP mechanisms, explaining its limited empirical relevance in the USD/CNY context (Cheung et al., 2007).
购买力平价 （PPP） 是最著名的观点之一 ，它认为应该调整汇率以使两个国家的相同商品篮子达到相同的价格。根据购买力平价的绝对版本，国内与国际价格水平的比率应与名义汇率相同。根据实证研究中更常使用的购买力平价的相对形式 ，两个国家之间的通货膨胀率差异相当于货币贬值或升值的速度。基本前提是，通过 贸易和资金流动的套利来保证跨境价格趋同。然而，随着时间的推移，购买力平价通常会保持得更好，不可贸易物 、市场摩擦和交易成本会带来短期变化。Froot 和 Rogoff （1995） 认为，偏离 PPP 可能会持续数年，尤其是在发展中或转型经济体。 就中国而言，对定价和资本流动的严格控制进一步抑制了 PPP 机制，这解释了它在美元/CNY 背景下的有限实证相关性（Cheung et al.， 2007）。

Uncovered Interest Parity (UIP), which posits a connection between interest rate differentials and anticipated fluctuations in exchange rates, is another important hypothesis. According to UIP, arbitrage possibilities in the global capital market should be reflected in the predicted appreciation (or depreciation) of a currency, which should balance the interest rate difference between two nations. In other words, when exchange rate risk is taken into account, investors should not care whether they invest in local or international assets. UIP makes the assumption that investors are risk-neutral and have perfect capital mobility. The "forward premium puzzle," which states that higher interest-rate currencies typically appreciate rather than devalue as UIP predicts, is a phenomena where empirical testing frequently refute UIP.
未覆盖利率平价 （UIP） 是另一个重要的假设，它假设利差与预期汇率波动之间存在联系。根据 UIP 的说法 ，全球资本市场的套利可能性应该反映在货币的预测升值（或贬值）上，这应该平衡两个国家之间的利率差异。换言之，当考虑到汇率风险时 ，投资者不应关心他们是投资于本地资产还是国际资产。 UIP 假设投资者是风险中性的，并且具有完美的资本流动性。“远期溢价之谜”指出，高利率货币通常会升值，而不是像 UIP 预测的那样贬值 ，这是实证检验经常反驳 UIP 的现象 。

Founded on the quantity theory of money, the monetary models of exchange rate influential factors were first created by Frenkel and Mussa (1976) and relate exchange rates to interest rates, the level of income, and the money supply. According to these ideas, exchange rates adapt to balance the supply and demand of money in each nation because prices are flexible and markets clear instantly. The flexible-price monetary model and the sticky-price model are two variations that take expectations and nominal rigidities into account. According to the monetary approach, lower real interest rates and higher inflation will cause currency depreciation in a nation with a larger money supply expansion than another.
汇率影响因素的货币模型建立在货币数量理论的基础上， 最早由 Frenkel 和 Mussa （1976） 创建，并将汇率与利率、 收入水平和货币供应量联系起来。根据这些想法，汇率会进行调整以平衡每个国家的货币供求关系，因为价格是灵活的，市场是即时明朗的。灵活价格货币模型和粘性价格模型是考虑预期和名义刚性的两种变体。根据货币方法，较低的实际利率和较高的通货膨胀将导致货币贬值，而货币供应量扩张幅度较大的国家将比另一个国家大。

However, seminal empirical studies, most notably Meese and Rogoff (1983), found that macroeconomic variables could not strongly explain variance in exchange rates. The literature also includes comparative works evaluating the forecasting performance of these models using real-time and revised data, such as those by Rossi (2013) and Cheung et al. (2005), who call for improved out-of-sample validation and robustness testing.
然而，开创性的实证研究，最著名的是 Meese 和 Rogoff （1983） 发现，宏观经济变量不能有力地解释汇率的差异 。 文献还包括使用实时和修订数据评估这些模型预测性能的比较工作，例如 Rossi （2013） 和 Cheung et al. （2005） 的比较工作，他们呼吁改进样本外验证和稳健性测试。

2.3. Empirical Evidence
2.3. 经验证明

The variables influencing exchange rate variations have been the subject of some relevant work (Beckmann et al., 2023). Halpem (1996) shows that external performance is the cause of real exchange rate dynamics. According to some research, demand shocks don't really matter because supply and nominal shock movements are what cause exchange rate swings (Kempa and Wilde, 2011). According to Alexius and Annika (2001), one of the elements that cannot be disregarded in determining the volatility of the real exchange rate is productivity improvement. Researchers have also discovered common (international) causes, meaning that genuine factors' influence on currency rate swings has grown in significance (Yepez and Dzikpe, 2022). Research has also demonstrated that supply, real demand, and nominal shocks are the main causes of exchange rate fluctuations (Wang et al., 2022). While the comparative values of nontraded commodities have negligible impacts on exchange rate movements, Bettendorf and Chen (2013) show that the comparable prices of traded items have an impact. Guo et al. (2021) come to the conclusion that long-term volatility in the RMB exchange rate is caused by capital inflows brought on by China's fast economic expansion. According to Wang et al. (2022), fluctuations in the price of oil have an effect on variations in the value of the currency in nations along the Belt and Road.
影响汇率变化的变量一直是一些相关工作的主题（Beckmann et al.， 2023）。 Halpem （1996） 表明，外部表现是实际汇率动态的原因。根据一些研究，需求冲击并不重要，因为供应和名义冲击变动是导致汇率波动的原因（Kempa 和 Wilde，2011 年）。 根据 Alexius 和 Annika （2001） 的说法， 在确定实际汇率的波动性时，不能忽视的要素之一是生产率的提高。研究人员还发现了常见的（国际） 原因，这意味着真正的因素对货币汇率波动的影响越来越重要（Yepez 和 Dzikpe，2022 年）。 研究还表明，供给、实际需求和名义冲击是汇率波动的主要原因（Wang et al.， 2022）。虽然非贸易商品的比较价值对汇率变动的影响可以忽略不计，但 Bettendorf 和 Chen （2013） 表明 ， 贸易物品的可比价格也有影响。Guo et al. （2021） 得出的结论是， 人民币汇率的长期波动是由中国经济快速扩张带来的资本流入引起的。根据 Wang et al. （2022） 的研究，石油价格的波动会影响“一带一路”沿线国家的货币价值变化。

The movement of exchange rates depends heavily on the level of difference between national inflation rates. When inflation remains low in a nation its currency value rises because domestic products become superior to foreign products overseas. Conversely, countries with higher inflation experience currency depreciation due to relatively higher prices for their goods (Beckmann and Czudaj, 2017; Ciccarelli, 2017). The difference between interest rates is also found to have impact on exchange rates because rising interest rates pull money across borders which enhances currency demand hence it leads to currency appreciation. A reduction in interest rates leads to currency depreciation as suggested by Kouwenberg et al. (2017). Moreover, the exchange rate will undergo changes based on the current account balances between two nations. Countries operating with trade deficits must deal with currency worth loss because their foreign currency requirements for imports increase. The appreciation of domestic currencies occurs when countries hold trade surpluses since buyers from abroad want to buy more domestic money (Rossi, 2013). Another determinant is the level of government debt. The exchange rates of countries having elevated public debt show depreciation when investors doubt inflation prospects or debt default risks emerge (Kouwenberg et al., 2017). Using quarterly data from 1970 to 2022, Sanusi et al. (2023) conduct a quantitative analysis that examines the link trade balance and, exchange rate, foreign direct investment, and economic growth. The findings also demonstrate that exports, as a part of the trade balance, and the exchange rate's performance both support economic development in the intended direction.
汇率的变动在很大程度上取决于国家通货膨胀率之间的差异水平。当一个国家的通货膨胀率保持在低位时，其货币价值就会上升，因为国内产品优于海外的外国产品。相反，由于商品价格相对较高，通货膨胀率较高的国家会出现货币贬值（Beckmann 和 Czudaj，2017 年 ; Ciccarelli，2017 年）。 利率之间的差异也被发现对汇率有影响 ， 因为利率上升会跨境拉动资金，从而增加货币需求，从而导致货币升值。 正如 Kouwenberg 等人（2017 年）所建议的，利率的降低与货币贬值有关 。此外， 汇率将根据两个国家之间的经常账户余额发生变化。贸易逆差国家必须处理货币价值损失，因为其进口的外汇需求增加。当国家持有贸易顺差时，本国货币就会升值，因为国外买家想要购买更多的国内货币（Rossi，2013）。 另一个决定因素是政府债务水平。 当投资者怀疑通胀前景或出现债务违约风险时，公共债务高企国家的汇率会贬值（Kouwenberg et al.，2017）。 使用 1970 年至 2022 年的季度数据，Sanusi 等人。 （2023） 进行定量分析，检查贸易平衡和汇率、外国直接投资和经济增长。研究结果还表明，作为贸易差额一部分的出口和汇率表现都支持经济朝着预期方向发展。

Economic growth is causally influenced by exchange rate performance, and vice versa, according to the Dumistrescu–Hurlin panel causality test. The goal of Etale and Ochuba's (2019) study was to assess how Nigerian output was affected by the trade balance and exchange rate performance. Multiple regression analysis and descriptive statistics were employed in the current study to assess the data gathered between 2000 and 2017. The latter demonstrated that the exchange rate had a significant impact on GDP based on data collected empirically. Given these realities, the country ought to take appropriate action and begin diversifying its economy with the sole goal of reducing crude oil exports. Studies like Husain et al. (2005) demonstrate the inverse association between economic growth and the fixed extent of the exchange rate. Most notably, a floating exchange rate system is more stable and strongly correlated with economic development, according to Husain et al. (2005). This is because the exchange rate system's flexibility benefits wealthier and more developed nations more. On the other hand, a fixed exchange rate system, although being the least stable and more vulnerable to risks during crises, lowers inflation and increases economic stability in emerging nations since they are less sensitive to the capital markets. On the other hand, many research, using a variety of data kinds, time periods, and estimating methods, fail to establish any solid proof that the exchange rate regime has an impact on economic growth. Regression was used to panel data from 18 industrialised nations between 1976 and 2001 by Huang and Malhotra (2005). The switching estimation approach was used by Domaç et al. (2004) for 22 distinct nations throughout a range of time periods.
根据 Dumistrescu-Hurlin 面板因果关系检验， 经济增长 受汇率表现的因果影响 ，反之亦然。Etale 和 Ochuba （2019） 研究的目标是评估尼日利亚产出如何 受到贸易平衡和汇率表现的影响。当前研究采用多元回归分析和描述性统计 来评估 2000 年至 2017 年间收集的数据。后者表明，根据实证收集的数据，汇率对 GDP 有重大影响。鉴于这些现实，该国应该采取适当的行动，开始实现经济多元化，其唯一目标是减少原油出口。 像 Husain 等人（2005 年）这样的人证明了经济增长与汇率的固定范围之间存在负相关。 最值得注意的是，根据 Husain 等人（2005 年）的说法，浮动汇率制度更稳定，并且与经济发展密切相关。这是因为汇率制度的灵活性使较富裕和较发达国家受益更多。另一方面，固定汇率制度虽然在危机期间最不稳定且更容易受到风险的影响，但由于新兴国家对资本市场的敏感度较低，因此可以降低通货膨胀并提高经济稳定性。 另一方面， 任何使用各种数据类型、 时间段和估计方法的研究都无法建立任何确凿的证据证明汇率制度对经济增长有影响。 Huang 和 Malhotra （2005） 对 1976 年至 2001 年间 18 个工业化国家的数据进行了回归分析 。 Domaç et al. （2004） 在一系列时间段内对 22 个不同的国家使用了转换估计方法 。

Meyer and Hassan (2020) reviewed the effect of exchange rate volatility concerning the South African government bond market. In the analysis, the author applied monthly data in the period extending between December 2018 and January 2000. To conduct this test they employed Johnsen cointegration test that identifies long-term relationship and, in turn, identifies the long-term impact of the exchange rate volatility on the frequency of the government bond market. They examined how the exchange rate regime affects South Africa international visitoration inflows between 1990 and 2020. They argued that the current float system of exchange rate promotes entrance of foreign tourists depending on the initial least square evaluation.
Meyer 和 Hassan （2020） 回顾了汇率波动对南非政府债券市场的影响。在分析中，作者应用了 2018 年 12 月至 2000 年 1 月期间的月度数据。为了进行这项测试，他们采用了 Johnsen 协整检验来识别长期关系，进而确定汇率波动对政府债券市场频率的长期影响。他们研究了汇率制度如何影响 1990 年至 2020 年间南非的国际游客流入。他们认为，目前的浮动汇率制度促进了外国游客的入境，这取决于最初的最小二乘评估 。

The factors of political security alongside total economic functioning prove fundamental for measuring exchange rates. Economic turbulence together with political instability in nations causes a depreciation of their currency because investors move toward secure investment opportunities. Countries achieving robust economic success alongside stable political systems normally draw foreign investment that enhances their currency value (de Grauwe & Kaltwasser, 2007). Finally, exchange rates are influenced by terms of trade which represent the relationship between export prices and import prices of a nation. Trade improvements resulting in export prices growing faster than import prices bring about currency appreciation as domestic goods gain more market demand (Cheung and Chinn, 2001; Jongen et al., 2008). Foreign exchange reserves indicate the ability of a central bank to meddle in the foreign exchange. In case of China, the large reserves have been seen as a strategic activity to avoid the appreciation of Yuan and sustain the export led growth (Zhou, 2009). Research findings (Obstfeld, Shambaugh, and Taylor, 2010) support the idea that build-up of reserves distorts exchange rate determination, because it overrides the market indicators. To the extent that one uses empirical evidence (Aizenman & Lee, 2007), they demonstrate that it is frequently possible to succeed intervention in the short run but develop the misalignments that compose over time. Thus, although reserves are not one of the primal determinants, another element major policy variable, which has mediated effects of fundamentals.
政治安全因素以及总体经济运作被证明是衡量汇率的基础。经济动荡加上各国的政治不稳定导致本国货币贬值，因为投资者转向安全的投资机会。在稳定的政治制度下取得强劲经济成功的国家通常会吸引外国投资，从而提高其货币价值（de Grauwe & Kaltwasser，2007 年）。最后，汇率受贸易条款的影响 ，贸易条款代表一个国家的出口价格和进口价格之间的关系。贸易改善导致出口价格增长速度快于进口价格，随着国内商品获得更多市场需求，货币升值（Cheung 和 Chinn，2001 年;Jongen et al.， 2008）。 外汇储备表明中央银行干预外汇的能力。就中国而言，大量外汇储备被视为避免人民币升值并维持出口主导增长的战略活动（周，2009）。研究结果（Obstfeld、Shambaugh 和 Taylor，2010 年）支持这样的观点，即储备的积累会扭曲汇率决定，因为它验证了市场指标。 只要 使用经验证据（Aizenman & Lee，2007），他们就会证明，短期干预的成功往往是可能的，但随着时间的推移，会出现错位。 因此， 尽管储备不是主要决定因素之一，但另一个因素是主要政策变量，它中介了基本面的影响。

2.4. Literature Gaps
2.4. 文献空白

While traditional models like PPP and IRP offer a theoretical basis for exchange rate movements, they often fail to explain behavior in economies with non-market exchange rate systems. Emerging empirical models that include a broader range of fundamentals such as current accounts and reserveshave improved explanatory power, but limitations remain. Particularly in China’s case, the literature shows that macroeconomic fundamentals alone are insufficient, and any effective model must incorporate policy discretion, capital controls, and institutional characteristics. As such, empirical studies that combine fundamental variables with regime-aware econometric modeling provide the most accurate reflection of exchange rate dynamics.
虽然 PPP 和 IRP 等传统模型为汇率变动提供了理论基础，但它们往往无法解释非市场汇率体系经济体的行为。包括更广泛的基本面 （如经常账户和准备金 ）的新兴实证模型提高了解释力，但局限性仍然存在。 特别是在中国的情况下，文献表明，仅靠宏观经济基本面是不够的， 任何有效的模型都必须考虑政策自由裁量权、资本控制和制度特征。因此，将基本变量与政权感知计量经济学模型相结合的实证研究提供了对汇率动态的最准确反映。

Chapter 3 Research Methods
第 3 章 研究方法

3.1. Research Design
3.1. 研究设计

The research focuses on positivist epistemology where the study will be quantitative in nature. Positivism presumes that the truth is objective- measurable, and that it is possible to identify causal nature of variables as a result of the empirical observation and statistical analysis (Bell, Bryman & Harley, 2018). This would be especially appropriate in macroeconomic research based on time-series statistics in which measurable variables like exchange rates, inflation, interest rates and trade balances offers empirical information to model.
该研究侧重于实证主义认识论，其中研究本质上是定量的。实证主义假定真理是客观可测量的，并且有可能 通过实证观察和统计分析来识别变量的因果性质（贝尔， 布赖曼和哈雷，2018）。这在基于时间序列统计的宏观经济研究中尤其适用，其中汇率、通货膨胀、利率和贸易差额等可衡量的变量为建模提供了实证信息。

The deductive technique in reasoning is adopted, wherein, the study starts with proven economic theories related to exchange rates as well as inflation and interest rates, in formulating the hypotheses. They then are experimentally tested on historical data. The objective is not only to narrate about exchange rate patterns but also to statistically identify whether the assumptions in theoretical macroeconomic fundamentals can explain variability in the USD/CNY exchange rate in a 25-year period.
采用推理中的演绎技术， 其中， 该研究从与汇率以及通货膨胀和利率相关的经过验证的经济理论开始 ，以制定假设。然后，根据历史数据对它们进行实验测试 。其目标不仅是叙述汇率模式，而且还要从统计上确定理论宏观经济基本面中的假设是否能够解释 25 年期间美元/人民币汇率的变化。

The explanatory quantitative strategy of the research is based on the determination of the causal relations between the macroeconomic indicators and the detailed exchange rate using USD and CNY. When one is to find out the level of influence of independent variables on a dependent variable, such studies are specially well-suited when the objective is to discover the strength and direction of said influences (Saunders, Lewis & Thornhill, 2019). There is a vast and widespread usage of quantitative research strategies in the location of international finance and macroeconomics, particularly in cases in which their need were focused in the determination of exchange rate movements. These methods also focus on objective measurement, numerical data and statistical methods in order to test hypothesis and test variables relationships. Since the commensurate inference of exchange rate economics is quite data-based and theory-based, the quantitative approach provides a strong framework of studying the impact of macroeconomic fundamentals which include interest rates, inflation, GDP growth and current account balances on currency values as time progresses. The dependent variable in this case is the yearly USD/CNY spot exchange rate, the independent variables have been the interest rates, inflation rates, GDP growth, current account balances and foreign exchange reserves. Econometric analysis based on the multiple linear regression will provide scientific operationalization of this strategy. Using this strategy, the study can determine whether the changes in macroeconomic fundamentals statistically explain changes in exchange rate and to which degree.
该研究的解释性定量策略基于使用美元和人民币确定宏观经济指标与详细汇率之间的因果关系。当人们要找出自变量对因变量的影响程度时，这样的研究特别适合于发现所述影响的强度和方向（Saunders， Lewis & Thornhill， 2019）。 在国际金融和宏观经济学的领域，定量研究策略被广泛而广泛地使用，特别是当它们的需求集中在确定汇率变动的情况下。这些方法还侧重于客观测量、数值数据和统计方法，以检验假设和检验变量关系。由于汇率经济学的相应推论是完全基于数据和理论的，因此定量方法提供了一个强大的框架来研究宏观经济基本面（包括利率、通货膨胀、GDP 增长和经常账户余额）随着时间的推移对货币价值的影响。 在这种情况下，因变量是美元/人民币的年度即期汇率， 自变量是利率、通货膨胀率、GDP 增长、经常账户余额和外汇储备。 基于多元线性回归的计量经济学分析将为该策略提供科学的实施。 使用这种策略，该研究可以确定宏观经济基本面的变化是否在统计学上解释了汇率的变化以及在多大程度上解释了汇率的变化。

3.2. Data Collection
3.2. 数据收集

The dataset consist of the exchange rate and macroeconomic variables from the United States and China. The primary variable of interest is the nominal bilateral exchange rate between the US dollar (USD) and Chinese yuan (CNY), which will be collected on a yearly basis to balance data availability with the macroeconomic variables. Macroeconomic variables will include inflation rates (measured by CPI), short-term interest rates (central bank policy rates), GDP growth rates, current account balances, and foreign exchange reserves. These variables are selected based on their theoretical relevance to exchange rate models and empirical usage in prior forecasting literature.
该数据集由来自美国和中国的汇率和宏观经济变量组成。感兴趣的主要变量是美元 （USD） 和人民币 （CNY） 之间的名义双边汇率，该汇率将每年收集 一次，以平衡数据可用性与宏观经济变量。 宏观经济变量将包括通货膨胀率（以 CPI 衡量）、短期利率（ 央行政策利率）、GDP 增长率、经常账户余额和外汇储备。这些变量是根据它们与汇率模型的理论相关性和先前预测文献中的实证用法来选择的 。

Data sources will include Bloomberg and Datastream for financial and exchange rate data, and FRED, the IMF’s International Financial Statistics (IFS), BIS, and central banks (the Federal Reserve and the People’s Bank of China) for macroeconomic indicators. The timeframe for analysis will span from 2000 to the most recent data available in 2024, capturing major global financial episodes including the global financial crisis, the US-China trade conflict, the COVID-19 pandemic, and post-pandemic recovery efforts.
数据源将包括 Bloomberg 和 Datastream 的金融和汇率数据，以及 FRED、国际货币基金组织的国际金融统计 （IFS）、BIS 和中央银行（ 美联储和中国人民银行）的宏观经济指标。分析的时间范围将从 2000 年到 2024 年的最新数据 ， 捕捉全球重大金融事件，包括全球金融危机、美中贸易冲突、COVID-19 大流行和大流行后的复苏努力。

3.3. Empirical Model
3.3. 经验模型

The study employs a Multiple Linear Regression model to test the relationship between the exchange rate and the selected macroeconomic variables. The model is specified as follows:
该研究采用多重线性回归模型来检验汇率与所选宏观经济变量之间的关系。模型指定如下：

+

Where:
哪里：

​ is the dependent variable: the annual USD/CNY exchange rate.
是因变量：美元/人民币的年汇率。

The βs are coefficients to be estimated.
β 是要估计的系数 。

εt​ is the error term capturing unexplained variation.
εt 是捕获无法解释的变化的误差项。

This model was selected because it enables simultaneous evaluation of multiple predictors, while controlling for collinearity and isolating the marginal effect of each independent variable on the dependent variable.
选择此模型 是因为它支持同时评估多个预测变量，同时控制共线性并隔离每个自变量对因变量的边际效应。

Guided by economic theory and previous empirical findings, the following hypotheses are formulated
在经济理论和以前的实证发现的指导下，提出了以下假设:

H1: There is a positive relationship between U.S. inflation and the USD/CNY exchange rate.
H1：美国通胀与美元/人民币汇率呈正相关。
H2: There is a negative relationship between China’s inflation and the USD/CNY exchange rate.
H2：中国的通货膨胀与美元/人民币汇率之间存在负相关关系。
H3: There is a positive relationship between the U.S. interest rate and the USD/CNY exchange rate.
H3：美国利率和 USD/CNY 汇率之间存在正相关关系。
H4: There is a negative relationship between China’s interest rate and the USD/CNY exchange rate.
H4：中国的利率和美元兑人民币汇率呈负相关。
H5: Higher U.S. GDP growth leads to a stronger dollar (lower USD/CNY).
H5：美国 GDP 增长加快导致美元走强（美元兑人民币走低）。

H6: Higher China GDP growth leads to a stronger Yuan (lower USD/CNY).
H6：中国 GDP 增长较高导致人民币走强（美元/人民币走低）。

H7: There is a negative relationship between the U.S. current account deficit and the USD/CNY exchange rate.
H7：美国经常账户赤字与 USD/CNY 汇率之间存在负相关关系。

H8: There is a negative relationship between China’s current account surplus and the USD/CNY exchange rate.
H8：中国经常账户顺差与美元兑人民币汇率呈负相关关系。

H9: There is a negative relationship between China’s foreign exchange reserves and the USD/CNY exchange rate
H9：中国的外汇储备与 USD/CNY 汇率之间存在负相关关系.

Chapter 4 Findings and Evaluation
第 4 章 调查结果和评估

4.1. Descriptive Analysis
4.1. 描述性分析

The figure below presents the changes of Chinese Yuan Renminbi (CNY) to U.S. Dollar (USD) from 2000 to 2024.
下图显示 2000 年至 2024 年人民币 （CNY） 兑美元 （USD） 的变化 。

Figure 1: Chinese Yuan Renminbi (CNY) to U.S. Dollar (USD) from 2000 to 2024 (Source: FRED, 2025)
图 1：2000 年至 2024 年人民币 （CNY） 兑美元 （USD） 的汇率 （来源：FRED，2025 年）

The chart shows the Chinese Yuan Renminbi (CNY) to U.S. Dollar (USD) spot exchange rate from around 2000 to mid-2024. From 2000 to 2005, the exchange rate was heavily pegged at about 8.28 CNY/USD. This reflects China’s fixed exchange rate policy under tight capital controls to maintain export competitiveness.
该图表显示了 2000 年左右至 2024 年年中人民币 （CNY） 兑美元 （USD） 的即期汇率。 从 2000 年到 2005 年，汇率严重固定在 8.28 元/美元左右。 这反映了中国在严格的资本管制下采取固定汇率政策，以保持出口竞争力。

In July 2005, China changed its exchange rate system to one that is more flexible. Yuan appreciated gradually at a rate of ~8.28 CNY/USD to ~6.83 CNY/USD. The shifts in the rate were conditioned by the fact that the United States together with the rest of the world demanded to decrease the trade imbalances along with the powerful export-driven development and huge trade surpluses on the part of China. But later in 2008, Yuan was re-pegged at the rate of 6.83 to stabilize exports and boost economic growth concerning the crisis of 2008. This move is also an indication of a conservative role played by Chinese policy makers in a global uncertainty. China was back to appreciation after the crisis and by 2014 it was at about 6.04. Through this also, one can learn how much the system of the managed float was closely controlled. Although the Yuan was officially being demarcated, the slowed rate and the rate at which the currency then appreciated was still dominated by government interventions more than market determined forces. This restricts the scope to which we can talk this period as a real liberalization of currency regime. It emerged as a minimalist and safety-first transition that would satisfy more the international stakeholders than the real relinquishment of control.
2005 年 7 月，中国将其汇率制度更改为更灵活的汇率制度。人民币以 ~8.28 CNY/USD 至 ~6.83 CNY/USD 的汇率逐渐升值。汇率的变化取决于美国与世界其他地区一起要求减少贸易失衡以及中国强大的出口驱动发展和巨额贸易顺差这一事实。 但在 2008 年晚些时候，人民币重新与 6.83 的汇率挂钩，以稳定出口并促进 2008 年危机的经济增长。此举也表明中国政策制定者在全球不确定性中扮演保守角色。危机后，中国重新升值，到 2014 年约为 6.04。通过这一点，人们还可以了解托管浮动的系统受到的严格控制程度。尽管人民币被正式划定，但汇率放缓和当时货币升值的速度仍然更多地受到政府干预的主导，而不是市场决定的力量。这限制了我们可以将这一时期称为货币制度真正自由化的范围。它以一种最低限度和安全第一的过渡出现，比起真正放弃控制权，它更能满足国际利益相关者的需求。

Lastly, the exchange rate has become very volatile in the period 2015 to 2019. This shift is part of the new beginning, as China is more open to the market influence and it has to reckon with new challenges, including capital outflows, stocks market volatility, and the consequences of the entire trade war between the U.S. and China.
最后，汇率在 2015 年至 2019 年期间变得非常不稳定。这种转变是新开始的一部分，因为中国对市场影响更加开放，它必须应对新的挑战，包括资本外流、股市波动以及美中之间整个贸易战的后果。

On examining inflation, the U.S. inflation rate has a mean of approximately 2.99 percent compared to that of China which has a higher average of 3 percent and with a higher volatility (standard deviation of 2.72 as compared to 1.83 of the U.S.). The severe minimum of inflation in China (-7.71 percent) underlines the instances of deflation, which could be connected to the internal corrections or the global shocks. Such a difference in inflation is vital since continuous inflation of a particular country as compared to another theoretically will cause depreciation of the currencies in that particular country according to the purchasing power parity (PPP). The overall average figures and yet high variability in the China data however point at an adjustment of the price levels structurally and policy-wise and not a stable inflation process.
在研究通货膨胀率时，美国的通货膨胀率平均值约为 2.99%，而中国的平均通货膨胀率更高，为 3%，波动性更高（标准差为 2.72，而美国为 1.83）。中国的严重最低通胀率 （-7.71%） 凸显了通货紧缩的情况，这可能与内部修正或全球冲击有关。这种通货膨胀差异至关重要，因为理论上，特定国家与另一个国家相比的持续通货膨胀将导致该特定国家的货币根据购买力平价 （PPP） 贬值。然而，总体平均数据以及中国数据的高可变性表明，价格水平在结构和政策上有所调整 ，而不是稳定的通胀过程。

Differentials of interest are also telling. The interest rate of United States averages at 2.91% but of China is even higher at 5.25%. Such sporadic interest rate may have been advantageous in attracting capital inflows into China when the financial regime was a freely mobile capital regime. Theoretically, this arbitrage opportunity has been restricted in practice however with capital controls and the managed exchange rate system. The margin highlights the Chinese practice of keeping high domestic rates as a means of controlling credit growth and inflation as well as make growth stimulative.
感兴趣的差异也很能说明问题。美国的平均利率为 2.91%，但中国的利率甚至更高，为 5.25%。当金融制度是自由流动的资本制度时，这种零星的利率可能有利于吸引资本流入中国。从理论上讲，这种套利机会在实践中受到资本管制和管理汇率制度的限制。该利润率凸显了中国保持高国内利率的做法，以此作为控制信贷增长和通胀以及刺激增长的手段 。

Economically, the rate of GDP growth in China is 7.68 percent with a standard deviation of 2.56 as opposed to 1.40 percent in the U.S. average. This huge growth difference is an indication of the fast-growing economic indicators in China due to its post-WTO boom, infrastructure construction, and catching up with the industry. But the fluctuations in the rate of growth in China (2. 09 to 10. 76) indicate exposure to any shock on world demand besides inefficiencies in the domestic economic field, that are present in form of housing crises or population decline related inefficiencies. The U.S. growth rate is much more tranquil and flat, which suggests that the economy is more mature and service-based, where change is slower.
从经济上讲，中国的 GDP 增长率为 7.68%，标准差为 2.56，而美国平均水平为 1.40%。这种巨大的增长差异表明，由于 WTO 后的繁荣、基础设施建设和赶上该行业，中国的经济指标正在快速增长。但中国增长率的波动（2. 09 到 10. 76）表明，除了国内经济领域的低效率之外，世界需求还受到了任何冲击，这些冲击以住房危机或与人口下降相关的低效率的形式存在。美国的增长率要平静和平稳得多，这表明经济更加成熟，以服务为基础，而变化较慢。

Finally, the current account data reinforces global imbalances. The U.S. current account deficit (% of GDP) ranges from -6% to -2.1%, consistent with its consumption-driven, import-reliant economy. China’s persistent surpluses and large reserves represent the mirror image, driven by exports, industrial policy, and cautious consumption patterns. Yet this model faces growing constraints as global demand slows, reshoring trends rise, and domestic rebalancing efforts encounter friction.
最后，经常账户数据加剧了全球失衡。美国经常账户赤字（ 占 GDP 的百分比）从 -6% 到 -2.1% 不等，与其消费驱动、依赖进口的经济一致。中国持续的盈余和庞大的外汇储备代表了由出口、产业政策和谨慎消费模式驱动的镜像。然而，随着全球需求放缓、回流趋势上升以及国内再平衡努力遇到摩擦，这种模式面临着越来越多的限制。

4.2. Correlation Analysis
4.2. 相关性分析

Correlation matrix provides very important insights of the relationship between the USD / CNY exchange rate and key macroeconomic variables. It is important to note that exchange rate is negatively and strongly correlated (r = -0.764) with the U.S current account balance. This implies that eventually (and when the spread between the U.S current account deficits is bigger), the USD will depreciate vis-a-vis the Yuan (or the Yuan appreciates). That concurs with the classical macroeconomics theory in which a currency weakens (naturally) as there are sustained deficits.
相关性矩阵提供了关于美元/人民币汇率与关键宏观经济变量之间关系的非常重要的见解。需要注意的是，汇率与美国经常账户余额呈负相关且密切相关 （r = -0.764）。 这意味着最终（当美国经常账户赤字之间的利差更大时），美元兑人民币将贬值（或人民币升值）。这与经典的宏观经济学理论一致，在这种理论中，货币会随着持续的赤字而（自然地）贬值。

The relationship is also positively associated with the growth of the GDP in China (r = 0.434), which is quite counterintuitive to an extent. Theorotically an upward rise in China would be able to boost the yuan, as there would be more investment and investment inflows. Nevertheless, such positive association can be attributed to policy based devaluations to promote export in high growth periods especially in the periods before 2015. On the other hand, the exchange rate has negligible relationships with inflation in both economies (US: r = 0.0985, China: r = 0.0952), which explains that difference in the price levels does not sufficiently explain exchange rate changes. Likewise, relationships with the growth of its U.S. GDP are practically insignificant (r = 0.0377) further confirming dominance of short term differences in growth not as the driving exchange rate factor here.
这种关系也与中国 GDP 的增长 （r = 0.434） 呈正相关，这在某种程度上是相当违反直觉的。 从理论上讲 ，中国的升值将能够提振人民币，因为会有更多的投资和投资流入。然而，这种正相关可以归因于基于政策的贬值，以促进高速增长时期的出口，尤其是在 2015 年之前的时期。 另一方面， 汇率与两个经济体的通货膨胀关系可以忽略不计（美国：r = 0.0985，中国：r = 0.0952），这解释了价格水平的差异并不能充分解释汇率变化。同样，与其美国 GDP 增长的关系实际上微不足道 （r = 0.0377），进一步证实了短期增长差异的主导地位 ，而不是这里的驱动汇率因素。

The interest rate in China has moderate positive correlation with the exchange rate (r = 0.2938) and thus this could also portray that the stronger rates in China correlates to a weak Yuan. This can be attributed to the exertion to curb monetary conditions in face of capital outflow or overheat therefore necessitating looser or overvalued currency positions.
中国的利率与汇率呈中等正相关 （r = 0.2938），因此这也可以说明中国的利率较高与人民币疲软相关。这可以归因于面对资本外流或过热时努力抑制货币条件，因此需要更宽松或高估的货币头寸。

Lastly, China’s foreign reserves show a mild positive correlation (r = 0.0652), suggesting that higher reserves coincide with mild Yuan depreciation, perhaps due to interventions or accumulation during times of external imbalance. Overall, the most significant and theoretically coherent relationship is with the U.S. current account, while other correlations suggest that China’s exchange rate is less driven by conventional macro fundamentals and more influenced by strategic interventions and policy objectives.
最后，中国的外汇储备呈轻微的正相关 （r = 0.0652），这表明较高的外汇储备与人民币的温和贬值相吻合， 这可能是由于外部失衡时期的干预或积累。总体而言，最重要且理论上连贯的关系是与美国经常账户的关系，而其他相关性表明，中国的汇率较少受传统宏观基本面的影响，而更多地受到战略干预和政策目标的影响。

4.3. Regression Analysis
4.3. 回归分析

The multiple regression output provides strong statistical insight into the factors influencing the USD/CNY exchange rate. The overall model is highly significant (F(8,16) = 12.33, p < 0.001) with an R-squared of 0.8608, meaning approximately 86% of the variation in the exchange rate is explained by the included macroeconomic variables. This suggests the model fits the data very well.
多元回归输出为影响 USD/CNY 汇率的因素提供了强大的统计见解。整个模型非常显著 （F（8,16） = 12.33，p < 0.001），R 平方为 0.8608，这意味着大约 86% 的汇率变化是由包含的宏观经济变量解释的。这表明该模型与数据拟合得非常好。

Importantly, the current account balance in the U.S. gives a high negative and significant coefficient (-0.7564, p < 0.001). This establishes the fact that the acceleration of U.S. current account deficit has a powerfully negative relationship with a weaker USD (or stronger Yuan), which is the case as per the external balance models. This is a sole variable that is statistically significant at conventional levels, which demonstrates its dominant role in explaining exchange rate developments.
重要的是，美国的经常账户余额给出了很高的负系数和显著系数 （-0.7564， p < 0.001）。这确立了这样一个事实，即美国经常账户赤字的加速与美元疲软（或人民币升值）具有强大的负相关关系，根据外部余额模型，情况就是如此。这是在传统水平上具有统计显著性的唯一变量，这表明它在解释汇率发展方面起着主导作用。

Other variables such as interest rate, inflation rate, GDP growth and foreign reserves are all found to show insignificant coefficient (p > 0.1) which implies that these factors when considered in isolation have no significant influence on the exchange rate in this specification and hence even though there is theoretical justification to include them, their coefficient cannot be given a significant value. As an example, both the growth of U.S. GDP and the growth of GDPs in China have negative coefficients and these imply that the higher-growth the better the Yuan, although the statistics are not significant (p = 0.094 and 0.144 respectively). In a similar manner, there is no significant relationship between the foreign exchange reserves and China, which means that reserves do indicate what the policy is doing but they may not necessarily be behind the expectation of the market that there will be a linear relationship between changes in reserves and changes in expectations regarding the exchange rate.
其他变量，如利率、通货膨胀率、GDP 增长和外汇储备，都被发现显示出微不足道的系数 （p > 0.1），这意味着这些因素在单独考虑时对本规范中的汇率没有显着影响，因此即使有理论上的理由包括它们，它们的系数也不能被赋予显着值。 例如，美国 GDP 的增长和中国的 GDP 增长都具有负系数，这意味着增长率越高，人民币越好，尽管统计数据并不显著（分别为 p = 0.094 和 0.144）。同样，外汇储备与中国之间没有显著关系，这意味着外汇储备确实表明了政策的作用，但它们可能不一定落后于市场的预期，即外汇储备的变化和对汇率的预期之间存在线性关系。

4.4. Evaluation of Findings
4.4. 结果评估

According to PPP, the exchange rate should adjust to offset inflation differentials between two countries. If a country’s inflation is higher than its trading partner’s, its currency should depreciate in the long run. However, in this study, neither inflation variable is statistically significant, and the signs are contrary to PPP predictions. For example, a negative coefficient for U.S. inflation implies that higher U.S. inflation strengthens the dollar (or weakens the Yuan), which is inconsistent with PPP. Similarly, China’s inflation has a negligible and insignificant effect. Thus, this result suggests that PPP does not hold in the short to medium term for the USD/CNY exchange rate, possibly due to China’s managed exchange rate regime, price rigidities, and capital controls that decouple the currency from domestic price movements. While the analysis found that inflation differentials between China and the United States were statistically insignificant, several empirical studies report contrasting results. Beckmann and Czudaj (2017), for instance, find that inflation differentials exert a meaningful influence on exchange rate dynamics across a broad cross-section of economies, particularly in the long run. Ciccarelli (2017) also emphasizes the inflation-exchange rate linkage in advanced and emerging economies. The divergence in findings may be attributable to China’s unique exchange rate management practices, where inflation signals are often neutralized through policy instruments. This inconsistency supports the view that, although inflation differentials may influence exchange rates elsewhere, they have limited explanatory power in China’s managed currency regime.
根据购买力平价，汇率应进行调整以抵消两国之间的通胀差异。如果一个国家的通货膨胀率高于其贸易伙伴的通货膨胀率，那么从长远来看，该国的货币应该会贬值。然而，在这项研究中， 两个通货膨胀变量都没有统计学意义，并且这些迹象与 PPP 预测相反。例如，美国通胀的负系数意味着美国通胀上升会使美元走强（或削弱人民币），这与购买力平价不一致。同样，中国的通货膨胀也影响不大 。因此， 这一结果表明，购买力平价在中短期内不适用于美元/人民币汇率，这可能是由于中国的受控汇率制度、价格刚性和资本管制使货币与国内价格走势脱钩。 虽然分析发现中美之间的通胀差异在统计上不显著，但几项实证研究报告了截然不同的结果。例如，Beckmann 和 Czudaj （2017） 发现，通胀差异对广泛经济体的汇率动态产生了有意义的影响，尤其是从长远来看 。 Ciccarelli （2017） 还强调了发达经济体和新兴经济体的通货膨胀与汇率的联系。研究结果的差异可能归因于中国独特的汇率管理做法，其中通胀信号通常通过政策工具来抵消。 这种不一致支持了这样一种观点，即尽管通胀差异可能会影响其他地方的汇率，但它们在中国的管理货币制度中的解释力有限。

Moreover, in terms of the Interest Rate Parity theory, IRP states that differences in interest rates between two countries should be offset by changes in exchange rates. A higher interest rate in one country should lead to depreciation of its currency due to expected capital outflows. Again, both coefficients are statistically insignificant, and their signs are inconsistent or too small to support IRP. The U.S. interest rate has almost no effect, while China's higher interest rate is weakly associated with a weaker Yuan, contrary to IRP expectations. Therefore, the lack of support for IRP reflects China’s capital account restrictions, which prevent arbitrage and capital mobility necessary for IRP to function. It also indicates that monetary policy differentials between the U.S. and China may not translate directly into exchange rate changes due to heavy intervention by the People's Bank of China. Moreover, the ineffectiveness of interest rate fluctuations in the USD/CNY regression does not correlate with the empirical study conducted by Kouwenberg et al. (2017), which determined a close connection between interest rates changes and cross-border flows, thus, influencing exchange rates in open economies. Nevertheless, just like in case of inflation, Guo et al. (2021) provide evidence that interest rate parity conditions are loosely implemented in the Chinese setting owing to capital controls. In their study, the long run movement of the exchange rate is blamed more on capital inflows following economic growth than arbitrage depending on interest differentials. This goes in line with the reasoning that in part closed financial systems the interest rates are not as dominant as the investment fueled flows.
此外，根据利率平价理论，IRP 指出，两国之间的利率差异应被汇率变化所抵消。由于预期的资本外流，一个国家的较高利率应会导致其货币贬值。 同样，这两个系数在统计上不显著，并且它们的符号不一致或太小，无法支持 IRP。美国利率几乎没有影响，而中国的较高利率与人民币贬值的关系较弱，这与 IRP 的预期相反。 因此， 对 IRP 的缺乏支持反映了中国的资本账户限制，这阻碍了 IRP 运作所需的套利和资本流动。它还表明，由于中国人民银行的大力干预，美国和中国之间的货币政策差异可能不会直接转化为汇率变化 。 此外， 美元/人民币回归中利率波动的无效性与 Kouwenberg 等人（2017 年 ）进行的实证研究无关 ，该研究确定了利率变化与跨境流动之间的密切联系，从而影响开放经济体的汇率。然而，就像通货膨胀一样，Guo et al. （2021） 提供的证据证明，由于资本管制，利率平价条件在中国的实施很松散 。在他们的研究中，汇率的长期变动更多地归咎于经济增长后的资本流入，而不是取决于利差的套利。 这与以下推理是一致的：在部分封闭的金融体系中，利率并不像投资推动的资金流动那样占主导地位。

The insignificant strength of the GDP growth on the USD/CNY exchange rate takes a different result with those reported in emerging economies in which the output performance has been empirically associated with valuation of the exchange rate. To give an example, Etale and Ochuba (2019) using the data of Nigeria conclude that exchange rate performance has a significant contribution to GDP via trade and investment channels. In the same way, Sanusi et al. (2023), in a study on a panel of countries in the southern part of Africa, also document an independent significant causal relationship between GDP and performance of the exchange rate. Conversely, Husain et al. (2005) found out that in fixed or managed states, GDP growth rate is unlikely to have any effect on the exchange rates except when it is accompanied by liberal capital movements and flexible monetary policies. The fact that the present research fits into this trend supports the opinion that the effect that economic growth has on currency value is strongly dependent on the system of the exchange rate that is implemented.
美元/人民币汇率的 GDP 增长强度微不足道，这与新兴经济体报告的结果不同，新兴经济体的产出表现在经验上与汇率估值相关联。举个例子，Etale 和 Ochuba （2019） 使用尼日利亚的数据得出结论，汇率表现通过贸易和投资渠道对 GDP 做出重大贡献。同样，Sanusi 等人（2023 年）在对非洲南部国家小组的研究中，也记录了 GDP 与汇率表现之间存在独立的重要因果关系。相反，Husain 等人（2005 年）发现，在固定或管理国家，GDP 增长率不太可能对汇率产生任何影响，除非它伴随着自由的资本流动和灵活的货币政策。本研究符合这一趋势的事实支持了这样的观点，即经济增长对货币价值的影响在很大程度上取决于所实施的汇率制度 。

Moreover, the statistically significant effect of the U.S. current account deficit on the USD/CNY exchange rate is in line with the findings advanced by Rossi (2013), who asserts that external imbalances especially habitual trade deficits are central agents of currency depreciation. The panel study conducted by Sanusi et al. (2023) also comes to a similar conclusion that exports are also a component of trade balance that affects the exchange rate and economic growth. Such parallels reinforce that the findings have added to the case that current account positions remain among the most empirically robust predictors of exchange rate dynamics, at least bilateral dynamics. On the other hand, a weak current account surplus of China in the regression is contrary to its expectations, though it can be read with Guo et al. (2021), who state that reserve accretion in China counteracted the role of surpluses in causing currency appreciation.
此外， 美国经常账户赤字对美元/人民币汇率的统计学显著影响与 Rossi （2013） 提出的发现一致，后者断言外部失衡，尤其是习惯性贸易赤字是货币贬值的核心因素。Sanusi 等人（2023 年）进行的小组研究也得出了类似的结论，即出口也是影响汇率和经济增长的贸易平衡的一个组成部分。这种相似之处强化了这一发现，即经常账户头寸仍然是汇率动态最可靠的实证预测指标之一，至少是双边动态。另一方面，回归中中国疲软的经常账户顺差与其预期相反，尽管它可以 与 Guo 等人（2021 年）一起解读，他们指出中国的储备增加抵消了盈余在导致货币升值中的作用。

Lastly, the regression output to indicate that the foreign exchange reserves role of China is not significant as opposed to numerous pieces of work that described the mediating effectiveness of the foreign exchange reserves in determining the exchange rates. Aizenman and Lee (2007) integrate quantitatively that in East Asian economies reserve build-up is an instrument to impact exchange rates in an indirect way. In the same way, Obstfeld, Shambaugh and Taylor (2010) advance that reserves, although they might not serve as fundamental drivers, they intertwine the association between fundamentals and an outcome of exchange rate by policy intervention. Zhou (2009) confines his arguments to China itself, claiming that the build-up of the reserves by the PBoC was a well-planned move against the appreciation of RMB. Consequently, the insignificance of reserves in the regression could be a fact of the impossibility to isolate their effect by treating them as a buffer or policy indicator instead of a direct economic variable.
最后， 回归输出表明中国的外汇储备作用并不重要，这与描述外汇储备在确定汇率方面的中介效果的大量工作相反。 Aizenman 和 Lee （2007） 定量地整合了在东亚经济体中，储备积累是一种 间接影响汇率的工具。同样 ，Obstfeld、Shambaugh 和 Taylor （2010） 推进了储备，尽管它们可能不是基本驱动力，但它们通过政策干预将基本面与汇率结果之间的关联交织在一起。周 （2009） 将他的论点局限于中国本身，声称中国人民银行增加储备是针对人民币升值的精心策划的举措 。因此，储备金在回归中的微不足道可能是一个事实，即不可能通过将储备金视为缓冲或政策指标而不是直接经济变量来孤立其影响。

Chapter 5 Conclusion
第 5 章 结论

5.1. Summaries of Findings
5.1. 调查结果摘要

This study set out to investigate the macroeconomic determinants of the Chinese Yuan Renminbi (CNY) to United States Dollar (USD) exchange rate from 2000 to 2024, with a particular focus on identifying which economic fundamentals influence currency fluctuations over time. Using annual time series data and a multiple linear regression model, the study assessed the influence of several independent variables including inflation rates, interest rates, GDP growth rates, current account balances, and foreign exchange reserves on the USD/CNY spot exchange rate.
本研究旨在调查 2000 年至 2024 年人民币 （CNY） 对美元 （USD） 汇率的宏观经济决定因素 ，特别侧重于确定哪些经济基本面会随着时间的推移影响货币波动。该研究使用年度时间序列数据和多元线性回归模型，评估了几个自变量的影响 ，包括通货膨胀率、利率、GDP 增长率、经常账户余额、 外汇储备对美元/人民币即期汇率的影响。

Results of the regression analysis provide valuable insights. The current account balance of the U.S. was the only statistically significant one (p< 0.01) and the coefficient value was -0.7564. It implies that an increase in U.S current account deficit (as a proportion of GDP) by 1 percentage point is correlated with a 0.76 increase in the exchange rate of USD/CNY, which means that USD depreciates. This observation is very emphatic to the BoP theory, thereby asserting that external imbalance of US is a major macroeconomic determinant of its currency value against the Chinese Yuan. The rest of the variables which included the inflation rates, interest rates, GDP growth and the foreign exchange reserves of China could not be significantly tested at the 10 percent level. Though coefficients of some of them had signs that were expected (such as the U.S. GDP growth had negative impacts on the exchange rate indicating stronger dollar), their coefficients were not found to be statistically significant which implies such relationships should be taken with caution. Low results of the inflation and interest rate variables mean that there is not much to support Purchasing Power Parity (PPP) and Interest Rate Parity (IRP) under the present bilateral environment. The most probable reason is a partly fixed exchange rate regime in China, which has the potential to decouple the exchange rates with short-run macroeconomic fundamentals.
回归分析的结果提供了有价值的见解。 美国的经常账户余额是唯一具有统计学意义的余额 （p< 0.01），系数值为 -0.7564。这意味着美国经常账户赤字（占 GDP 的比例）增加 1 个百分点与美元/人民币汇率增加 0.76 相关，这意味着美元贬值。这一观察结果非常强调国际收支平衡理论，从而断言美国的外部失衡是其货币对人民币价值的主要宏观经济决定因素。 其余的变量，包括中国的通货膨胀率、利率、GDP 增长和外汇储备 ，无法在 10% 的水平上进行重大测试。尽管其中一些系数具有预期的迹象（例如美国 GDP 增长对汇率产生负面影响，表明美元走强），但它们的系数没有发现具有统计学意义，这意味着应谨慎对待这种关系。通货膨胀和利率变量的低结果意味着在目前的双边环境下，没有太多支持购买力平价 （PPP） 和利率平价 （IRP） 的东西。最可能的原因是中国的部分固定汇率制度，这有可能使汇率与短期宏观经济基本面脱钩。

Moreover, the foreign exchange reserves of China that in most cases had been used in stabilization of its currency were found not to be significant in the model. This can be likely to be because management of the reserves is complicated and reactive and not necessarily correlated directly to changes in spot exchange rate. Alternatively, the insignificance could be due to the problem of multicollinearity or measurement constraint since reserve accumulation is correlated with other macro economic indicators like trade balances, and capital flow.
此外，在大多数情况下用于稳定其货币的中国外汇储备 在该模型中被发现并不重要。这可能是因为储备的管理复杂且反应迟钝 ， 不一定与即期汇率的变化直接相关。或者，微不足道可能是由于多重共线性或测量约束的问题，因为储备积累 与其他宏观经济指标（如贸易差额和资本流动 ）相关 。

Overall, the findings suggest that while many standard macroeconomic variables have some theoretical relevance to exchange rate determination, the U.S. current account balance stands out as the most powerful and statistically valid predictor of USD/CNY exchange rate movements over the past two decades. This underscores the centrality of external imbalances in driving exchange rate trends in a world where monetary policy and inflation dynamics are often dampened by institutional controls, especially in the Chinese context.
总体而言，研究结果表明，虽然许多标准宏观经济变量与汇率确定具有一定的理论相关性，但美国经常账户余额是过去二十年来美元兑人民币汇率变动最强大且统计有效的预测指标。这突显了外部失衡在推动汇率趋势方面的核心作用，因为货币政策和通胀动态经常受到制度控制的抑制，尤其是在中国的情况下。

5.2. Research Limitations
5.2. 研究限制

Notwithstanding the fact that the study comes with significant findings, it has its shortcomings. The first significant weakness is the frequency of data and the sample size. Only 25 observations were available because the yearly data of 2000 to 2024 were used in the analysis. This small sample restricts statistics power of regression model and lowers capacities to identify small but significant effects. The superior resolution (e.g. quarterly or monthly) would lead to more accuracy and may reflect shorter-term dynamics that average annual numbers can conceal.
尽管该研究有重要的发现，但它也有其缺点。第一个显著的弱点是数据的频率和样本量。由于分析中使用了 2000 年至 2024 年的年度数据，因此只有 25 个观察结果可用。这个小样本限制了回归模型的统计功效，并降低了识别微小但显著效应的能力。卓越的分辨率（例如每季度或每月）将导致更高的准确性，并可能反映平均年度数字可以隐藏的短期动态。

Secondly, the research employed the linear regression model that supposes an adverse symmetrical correlation among variables. But the determination of the exchange rate tends to be of non-linear nature and can be subject to thresholds, regime changes and sporadic speculation. As an example, in case of financial crisis or sudden change of policy (like the mini-devaluation in 2015 or the COVID-19 pandemic), reactions could be out of proportion and non-linear in the market. In more detail, future studies might consider nonlinear models, structural break tests or time-varying parameter regressions in order to recognize such dynamics.
其次，该研究采用了线性回归模型，该模型假设变量之间存在负对称相关性。 但汇率的确定往往是非线性的，可能会受到阈值、政权更迭和零星投机的影响。例如，在金融危机或政策突然变化（如 2015 年的小型贬值或 COVID-19 大流行）的情况下，市场的反应可能不成比例且非线性。更详细地说，未来的研究可能会考虑非线性模型、结构断裂测试或时变参数回归，以识别此类动力学。

A third weakness of this research is the lack of crucial variables on capital accounts: foreign direct investment (FDI), portfolio flows and controls of capital. These contribute greatly to the dynamics of the exchange rate particularly in the Chinese situation where government manipulates the cross border capital flows. This omission of the variables could result in the model being biased with omitted variables that may skew the coefficients of the other variables included.
这项研究的第三个弱点是缺乏资本账户的关键变量：外国直接投资 （FDI）、投资组合流动和资本控制。这些对汇率的动态有很大贡献，尤其是在政府纵跨境资本流动的中国情况下。这种变量的省略可能会导致模型 与省略的变量产生偏差，这可能会扭曲所包含的其他变量的系数。

Also, the research study cannot directly explain the interventions of the exchange rates by the People Bank of China, as well as the political and geopolitical shocks such as the U.S-China trade tension which are crucial issues that have had considerable effect on bilateral exchange rates. This is very challenging to measure but it cannot be ignored as a conceptual shortcoming considering that the exchange rate plays a strategic role in Chinese industrial policy.
此外 ，该研究无法直接解释中国人民银行对汇率的干预，以及政治和地缘政治冲击，如美中贸易紧张局势，这些都是对双边汇率产生重大影响的关键问题。这是非常具有挑战性的衡量，但 考虑到汇率在中国产业政策中起着战略作用 ，它不能作为一个概念上的缺陷而被忽视 。

Lastly, the study is inherently bilateral and focused solely on the USD/CNY rate. This ignores multilateral exchange rate pressures and interactions with other currencies like the euro or yen, which also influence global capital flows and investor behavior. The findings may not generalize to broader exchange rate frameworks or other currency pairs.
最后，该研究本质上是双边的，仅关注美元/人民币汇率。这忽略了多边汇率压力和与欧元或日元等其他货币的互动，这些也影响着全球资本流动和投资者行为。这些发现可能无法推广到更广泛的汇率框架或其他货币对。

5.3. Suggestions for Future Research
5.3. 对未来研究的建议

Based on the findings and the shortcomings of this research, it is possible to propose a number of suggestions on how future research on this topic should be conducted. First of all, a higher frequency of data (such as monthly or quarterly) should be used in future studies, which would enable an improvement in the level of observations and hence the method used in the econometric estimation as more sound measures such as Vector Autoregression analysis, cointegration and Granger causality analysis. Dynamic interactions would be better picked up and long-run equilibrium relations between variables would be better represented through these methods. Second, the addition of capital flow variables, including the net FDI, portfolio inflows and openness of the capital account would give a more complete picture of the factors which influence the USD/CNY exchange rate. Due to the gradual liberalization of the Chinese capital account, monitoring of such variables may prove to be useful in providing an answer to turning points in the exchange rate.
根据这项研究的发现和缺点，可以就未来如何进行该主题的研究提出一些建议 。首先 ， 在未来的研究中应该使用更高频率的数据（例如每月或每季度），这将能够提高观测水平，从而提高计量经济学估计中使用的方法 ，作为更合理的措施，例如向量自回归分析、 协整和格兰杰因果关系分析。通过这些方法，可以更好地理解动态交互作用，并且更好地表示变量之间的长期均衡关系。 其次，增加资本流动变量，包括净 FDI、投资组合流入和资本账户开放性，将更全面地了解 影响美元兑人民币汇率的因素。由于中国资本账户的逐步自由化，对这些变量的监控可能有助于为汇率的转折点提供答案。

Third, other researchers can consider potential modeling of the intervention behavior by the PBoC possibily in the form of intervention proxies (including reserves change, sterilization coefficients) and dummy variables of known interventional periods. This modeling has the potential to explain how discretionary policy in the form of counter-pressures or augmentations. Also, it is recommended that in the future studies, we must test nonlinearities, involving structure breaks, through threshold models or Markov-switching models to allow consideration of various policy regimes (pre-2005 peg, post-2005 managed float, 2015 reform). These instruments would assist in the definition of a regular market pattern and extraordinary times of intervention or crisis.
第三， 其他研究人员可以考虑 PBoC 对干预行为的潜在建模可能在干预代理（包括 储备变化、灭菌系数）和已知干预期的虚拟变量的 m 中。这种建模有可能解释自由裁量政策如何以反压力或增强的形式出现。 此外，建议在未来的研究中，我们必须通过阈值模型或马尔可夫转换模型来测试非线性，涉及结构断裂，以允许考虑各种政策制度（2005 年之前的挂钩、2005 年后的管理浮动、2015 年的改革）。这些工具将有助于定义常规市场模式和干预或危机的特殊时期。

Lastly, it would also be useful to replicate the model as a multilateral in terms of the use of effective exchange rate indices or panel data about other main currencies. This would enable the comparison of determinants of the various exchange rate regimes and effectively focus on the international environment of exchange rate changes.
最后， 在使用有效汇率指数或其他主要货币的面板数据方面，将模型复制为多边模型也很有用。这将有助于比较各种汇率制度的决定因素，并有效地关注汇率变化的国际环境。

References
引用

Beckmann, J., & Czudaj, R. (2017). Exchange rate expectations since the financial crisis: Performance evaluation and the role of monetary policy and safe haven. Journal of International Money and Finance, 74(C), 283–300.
贝克曼，J.，和 Czudaj，R.（2017 年）。金融危机以来的汇率预期：绩效评估以及货币政策和避风港的作用。 国际货币与金融杂志，74（C），283-300。

Cheung, Y. W., & Chinn, M. D. (2001). Currency traders and exchange rate dynamics: A survey of the US market. Journal of International Money and Finance, 20(4), 439–471.
Cheung， Y. W.， & Chinn， M. D. （2001 年）。 外汇交易商和汇率动态：美国市场调查。 国际货币与金融杂志，20（4），439-471。

Cheung, Y. W., Chinn, M. D., & Pascual, A. G. (2005). Empirical exchange rate models of the nineties: Are any fit to survive? Journal of International Money and Finance, 24(7), 1150–1175.
Cheung， Y. W.， Chinn， M. D.， & Pascual， A. G. （2005 年）。 90 年代的经验汇率模型：任何适合生存的模型吗？ 国际货币与金融杂志，24（7），1150-1175。

Ciccarelli, M., García, J. A., & Montes-Galdón, C. (2017). Unconventional monetary policy and the anchoring of inflation expectations (ECB Working Paper Series No. 1995). European Central Bank. https://doi.org/10.2866/262979
Ciccarelli， M.， García， J. A.， & Montes-Galdón， C. （2017）。 非常规货币政策和通胀预期的锚定 （欧洲央行工作论文系列第 1995 号）。欧洲中央银行。https://doi.org/10.2866/262979

de Grauwe, P., & Kaltwasser, P. (2007). Modeling optimism and pessimism in the foreign exchange market. CESifo Working Paper Series No. 1962
deGrauwe， P.， & Kaltwasser， P. （2007）. 模拟外汇市场的乐观和悲观情绪。CESifo 工作论文系列第 1962 号

Frenkel, J. A., & Mussa, M. L. (1976). The monetary approach to the exchange rate: Some empirical evidence. Scandinavian Journal of Economics, 78(2), 200–224.
Frenkel， J. A.， & Mussa， M. L. （1976 年）。 汇率的货币方法：一些实证证据。 斯堪的纳维亚经济学杂志，78（2），200-224。

Jongen, R., Verschoor, W. F. C., & Wolff, C. C. P. (2008). Foreign exchange rate expectations: Survey and synthesis. Journal of Economic Surveys, 22 (1), 140-165
Jongen， R.， Verschoor， W. F. C.， & Wolff， C. C. P. （2008）.Foreign 汇率预期：调查和综合 。 经济调查杂志 ， 22 （1）， 140-165

Kouwenberg, R., Markiewicz, A., Verhoeks, R., & Zwinkels, R. C. J. (2017). Model uncertainty and exchange rate forecasting. Journal of Financial and Quantitative Analysis, 52(1), 341–363.
Kouwenberg， R.， Markiewicz， A.， Verhoeks， R.， & Zwinkels， R. C. J. （2017）.对不确定性和汇率预测进行建模。 金融与定量分析杂志，52（1），341-363。

Meese, R., & Rogoff, K. (1983). Empirical exchange rate models of the seventies: Do they fit out of sample? Journal of International Economics, 14(1–2), 3–24.
Meese， R.， & Rogoff， K. （1983 年）。 70 年代的实证汇率模型：它们是否适合样本？ 国际经济学杂志，14（1-2），3-24。

Rossi, B. (2013). Exchange rate predictability. Journal of Economic Literature, 51(4), 1063–1119.
罗西，B.（2013 年）。 汇率可预测性。 经济文献杂志，51（4），1063-1119。