金融學專業(yè)外文翻譯---巴西股票價格與匯率之間關(guān)系的實證分析

1、<p>　　本科畢業(yè)論文外文原文</p><p>　　外文題目：THE DYNAMIC RELATIONSHIP BETWEEN STOCK PRICES AND EXCHANGE RATES: EVIDENCE FOR BRAZIL </p><p>　　出 處：International Journal of

2、 Theoretical and Applied Finance </p><p>　　作 者：BENJAMIN M. TABAK </p><p>　　This paper studies the dynamic relationship between stock pri

3、ces and exchange rates in the Brazilian economy. We use recently developed unit root and cointegration tests, which allow endogenous breaks, to test for a long run relationship between these variables. We performed linea

4、r, and nonlinear causality tests after considering both volatility and linear dependence. We found that there is no long run relationship, but there is linear Granger causality from stock prices to exchange rates, in li&

5、lt;/p><p>　　Keywords: Stock prices; exchange rates; bivariate causality; nonlinear causality.</p><p>　　1. Introduction</p><p>　　The literature that studies the relationship between exc

6、hange rates and stock prices is far from conclusive. There are two main theories that relate these financial markets. The first is the traditional approach, which concludes that exchange rates should lead stock prices. T

7、he transmission channel would be exchange rate fluctuations which affect firm’s values through changes in competitiveness and changes in the value of firm’s assets and liabilities, denominated in foreign currency, ultima

8、tely</p><p>　　Alternatively, changes in stock prices may influence movements in exchange rates via portfolio adjustments (inflows/outflows of foreign capital). If there were a persistent upward trend in stoc

9、k prices, inflows of foreign capital would rise. 1Even firms that are not internationally integrated (low ratio of exports and imports to total sales and a low proportion of foreign currency-denominated assets and liabil

10、ities) may be indirectly affected.</p><p>　　However, a decrease in stock prices would induce a reduction in domestic investor’s wealth, leading to a fall in the demand for money and lower interest rates, cau

11、sing capital outflows that would result in currency depreciation. Therefore, under the portfolio approach, stock prices would lead exchange rates with a negative correlation.</p><p>　　In January 1999, Brazil

12、 abandoned the crawling peg and target zone regimes and adopted a floating exchange rate.2 From January 14 to March 3, the Brazilian Real depreciated drastically, 49.51%. The BOVESPA Index (the S?ao Paulo Stock Exchange

13、Index, the most important stock index in the country) increased 4.097 points in the same period (59.34% rise). This effect on the domestic stock index is very different from that observed in Asian economies at the start

14、of the Asian crisis. Therefore, the B</p><p>　　The rapid increase of the stock index could have occurred because economic worldwide believed that the currency was overvalued, and that depreciation would lead

15、 to an increase in firm competitiveness, enhancing exports and raising profits. Moreover, many firms that comprise the stock index have American DepositoryReceipts (ADR); these stock prices would respond almost immediate

16、ly through arbitrage mechanisms, since, with the rapid depreciation, domestic traded stockswould be very cheap vis-a-vis</p><p>　　We analyze the dynamics between the stock index and the exchange rate using l

17、inear, and nonlinear, Granger causality tests. We employ series filtered for volatility and linear dependence when performing the nonlinear causality tests. We make use of newly developed unit root and cointegration test

18、s, which allow endogenous breaks, to test for a long-run equilibrium relationship between these variables. Furthermore, we use impulse response functions to test the validity of both the traditional and </p><p

19、>　　This paper is organized as follows. In the next section, we present a brief literature review and the main findings in developed and emerging countries. Section 3 presents the data and methodology employed. Section

20、 4 shows the empirical evidence for the interdependencies between stock prices and exchange rates in Brazil. Section 5 concludes the paper and gives some directions for further research.</p><p>　　2. Literatu

21、re Review</p><p>　　The relationship between exchange rates and stock prices is of great interest to many academics and professionals, since they play a crucial role in the economy. Nonetheless, results are s

22、omewhat mixed as to whether stock indexes lead exchange</p><p>　　rates or vice versa and whether feedback effects (bi-causality) even exist among these financial variables.</p><p>　　Campa et al.

23、 [11] studied the credibility of the crawling peg and target zone (maxiband) regimes and have a nice description of the period prior to the maxi-devaluation of the Real in 1999. Dynamic Relationship Between Stock Prices

24、and Exchange Rates 1379</p><p>　　Aggarwal [4] argued that changes in exchange rates provoke profits or losses in the balance sheet of multinational firms, which induces their stock prices to change. In this

25、case, exchange rates cause changes in stock prices (traditional approach).</p><p>　　Dornbusch [14] and Boyer [10] presented models suggesting that changes in stock prices and exchange rates are related by ca

26、pital movements. Decreases in stock prices reduce domestic wealth, lowering the demand for money and interest rates, inducing capital outflows and currency depreciation.</p><p>　　Bahmani-Oskooee and Sohrabia

27、n [6] analyzed the relation between stock prices and exchange rates in the US economy. They found no long-run relationship among these variables, but a dual causal relationship in the short-run using Granger [16] causali

28、ty tests.3 Amihud [5] and Bartov and Bodnar [7] found that lagged, and not contemporaneous, changes in US dollar exchange rates, explain firms current stock returns.</p><p>　　Ratner [29] applied cointegratio

29、n analysis to test whether US dollar exchange rates affect US stock prices, using monthly data from March 1973 to December 1989. His results indicated that the underlying long-term stochastic properties of the US stock i

30、ndex and foreign exchange rates are not related, since the null of no cointegration could not be rejected, even when dividing the sample into sub-periods.</p><p>　　Ajayi and Mougou′e [3] analyzed the relatio

31、nship between stock prices and exchange rates in eight advanced economies (Canada, France, Germany, Italy, Japan, the Netherlands, the United Kingdom and the United States).4 Using an error correction model, they found s

32、ignificant short and long run feedback between these two variables.</p><p>　　Abdalla and Murinde [1] investigated interactions between exchange rates and stock prices in India, Korea, Pakistan, and the Phili

33、ppines. Using monthly observations in the period from January 1985 to July 1994. Within an error correction model framework, they found evidence of unidirectional causality from exchange rates to stock prices in all coun

34、tries, except for the Philippines. There, they found that stock prices Granger influence exchange rates.</p><p>　　Ong and Izan [28] used weekly data of “spot and 90-day forward” exchange rates for Australia

35、and the G-7 countries and “spot and 90-day forward” futures prices for equity prices in Australia, Britain, France and the US, during the period from October 1986 to December 1992. They were unable to find a significant

36、relationship</p><p>　　between equity and exchange rate markets. They suggested that the use of daily data (or even intra-day) could improve their empirical results.</p><p>　　Ajayi et al. [2] use

37、d daily data and reported that causality runs from the stock market to the currency market in Indonesia and the Philippines, while in Korea it runs in the opposite direction. No significant causal relation is observed in

38、 They use the S&P 500, the effective exchange rate, and monthly data over the period from July 1973 to December 1988. Their sample runs from April 1985 to July 1991. Hong Kong, Singapore, Thailand, or Malaysia. Howev

39、er, in Taiwan, they detected bi-directional </p><p>　　Granger et al. [17] found strong feedback relations between Hong Kong, Malaysia, Thailand and Taiwan. They used daily data and their sample period starte

40、d January 3, 1986 and finished June 16, 1998. Furthermore, they found that the results are in line with the traditional approach in Korea, while they agree with the portfolio approach in the Philippines.</p><p

41、>　　Nieh and Lee [26] found no significant long-run relationship between stock prices and exchange rates in G-7 countries, using both the Engle-Granger and Johansen’s cointegration tests.6 Furthermore, they found ambig

42、uous, and significant, shortrun relationships for these countries. Nonetheless, in some countries, both stock indexes and exchange rates may serve to forecast the future paths of these variables. For example, they found

43、that currency depreciation stimulates Canadian and UK stock market</p><p>　　In general, empirical findings suggest that there are no long-run equilibrium relationships between these two financial variables (

44、exchange rates and stock prices) in most countries. However, many studies have found that these variables have “predictive ability” for each other, although the direction of causality seems to depend</p><p>

45、　　on specific characteristics of the country analyzed. To the best of our knowledge, this is the first paper that addresses this issue in the Brazilian economy.</p><p>　　3. Data and Methodology</p>&l

46、t;p>　　The data, obtained from Bloomberg, consists of 1,922 observations, from August 1, 1994 to May 14, 2002, of daily closing prices in the S?ao Paulo Stock Exchange Index (IBOVESPA) and foreign exchange rate (units

47、of Real per US dollar). We use daily data since the use of monthly data may not be adequate to capture the effects of capital movements</p><p>　　1、Unit roots</p><p>　　We used the Augmented Dicke

48、y and Fuller [13] (ADF) test for unit roots, using both a trend and an intercept. In general, an ADF(p) model is given by</p><p>　　ΔXt = α + (1 ? φ)Xt?1 + γt +ΣβiΔXt?i + εt.(3.1)</p><p>　　The Ba

49、yesian Schwarz Information criterion was used to choose the order of lags (p) in Eq. (3.1). Furthermore, we imposed an additional requirement, that the resulting model has white noise residuals. If the resulting model ha

50、s serial correlation, the order of lags is augmented until residuals with no serial correlation are obtained.</p><p>　　Since the failure to reject the null of a unit root may be due to the low power of unit

51、root tests against stationary alternatives, Kwiatkowski, et al. [22] proposed a test where the null is stationary and the alternative is a unit root. This test is given by</p><p>　　KPSS =1/T2ΣS2t/S2(L)</

52、p><p>　　Where ST=Σei (t=1，2，3，……T)</p><p>　　And S2=1/TΣ(e2t)+2/TΣ(1-S/(L+1))Σ(etet-s)</p><p>　　The residuals are given by the e_i s, T is the number of observ

53、ations and L is the laglength.</p><p>　　譯 文：巴西股票價格與匯率之間關(guān)系的實證分析</p><p>　　本文研究在巴西經(jīng)濟中股價與匯率的動態(tài)關(guān)系，我們使用單位根檢驗和協(xié)整關(guān)系檢驗來研究主要變量之間的長期關(guān)系。同時在考慮波動性和線性關(guān)系的基礎(chǔ)上，我們還使用線性因果關(guān)系檢驗和非線性因果關(guān)系檢驗。通過檢驗，我們發(fā)現(xiàn)，兩者之間沒有任何長期關(guān)系，但按照

54、投資組合的方法可以得出股價與匯率之間是呈負相關(guān)的，股票價格對匯率存在線性格蘭杰因果關(guān)系。此外，我們還發(fā)現(xiàn)按照傳統(tǒng)的方法發(fā)現(xiàn)匯率對股票價格存在非線性格蘭杰因果關(guān)系。我們認為，這些發(fā)現(xiàn)對國際投資者和政策制定者有著實際的用處。</p><p>　　關(guān)鍵詞：股票價格，匯率，二元因果關(guān)系，非線性因果關(guān)系</p><p><b>　　1、介紹</b></p><

55、;p>　　本文所用的文獻中主要是通過遠離定律研究匯率與股價之間的關(guān)系。它涉及兩個主要的金融市場的理論。首先是傳統(tǒng)做法的理論，它認為匯率變動會導致股票價格的變動。匯率波動對股權(quán)價值的的影響是通過以外幣計價的公司的資產(chǎn)和負債的價值的變化最終引起公司的利潤的改變而實現(xiàn)的。而另一種方法是投資組合的方法理論，在投資組合的方法中通過調(diào)整利率使得股票價格的變化可能對匯率的變化帶來影響。如果保持股票價格持續(xù)上升的趨勢，將會導致國外資本的流入的增加

56、。而在股票價格下跌的時候，會引起國內(nèi)投資者的資產(chǎn)的外流，導致貨幣需求和利率的下降，而資本的外流使得國內(nèi)貨幣的貶值。因此，在投資組合的方法下，股票價格和匯率之間是呈負相關(guān)。</p><p>　　1999年1月，巴西放棄了固定匯率制度，而選擇使用浮動匯率制度。在1月14日至3月3巴西貨幣大幅貶值了49.51%，巴西證券交易所指數(shù)在同一時期上升了4.097個百分點（即上升了59.34%）。亞洲經(jīng)濟學家從巴西股價指數(shù)變化

57、對國內(nèi)經(jīng)濟的產(chǎn)生了重大的影響中觀察到了這是亞洲經(jīng)濟危機的開始。因此，巴西的情況為我們提供了一個學習股票價格與匯率動態(tài)關(guān)系的好機會。</p><p>　　巴西股票指數(shù)的快速增長的基礎(chǔ)是經(jīng)濟的快速增長，在全世界都認為貨幣價值是被高估時，貨幣的貶值將會給企業(yè)提供更多的出口量和利潤，增加企業(yè)的競爭力。此外，許多公司的股票指數(shù)的構(gòu)成包括美國存托憑證收據(jù)（ADR），因此在面對貨幣貶值的時候這些股票價格使用套利機制幾乎會立即做

58、出反應(yīng)，使得國內(nèi)交易的股票的價格在面對美國存托憑證收據(jù)時會變得很廉價。</p><p>　　我們通過對股票價格與匯率之間的線性檢驗，非線性檢驗及格蘭杰因果關(guān)系檢驗對兩者之間的關(guān)系進行分析。我們對有波動的數(shù)據(jù)采用線性和非線性因果關(guān)系檢驗的方法進行檢驗。我們采用新開發(fā)的單位根和協(xié)整檢驗，用來測試這些變量之間的長期均衡關(guān)系。此外，我們還使用脈沖響應(yīng)函數(shù)來測試傳統(tǒng)和組合的兩種方法對匯率與股價指數(shù)兩者之間關(guān)系的研究結(jié)果。&

59、lt;/p><p>　　本文敘述如下，在下一節(jié)中，我們提出了一個簡短的主要針對對發(fā)達國家和新興國家的研究結(jié)果的分析。第三節(jié)，介紹了數(shù)據(jù)和方法的應(yīng)用。第四節(jié)給出了對巴西的股價和匯率之間的相互關(guān)系的實證結(jié)果。第五節(jié)總結(jié)本文，并給出了結(jié)論和進一步研究方向。</p><p><b>　　2、文獻</b></p><p>　　許多學者和專業(yè)人士對匯率和股票價

60、格之間的關(guān)系非常感興趣，而且他們的研究在經(jīng)濟發(fā)展中發(fā)揮了關(guān)鍵作用。然而，研究的結(jié)果包括多種，股價指數(shù)與匯率之間可能存在股價指數(shù)變動導致匯率的變動，或者反之匯率的變化導致股價指數(shù)的變化，或者兩者是雙向的關(guān)系。</p><p>　　Aggarwal認為，匯率變動引起跨國公司的資產(chǎn)負債表上利潤或虧損的變動，從而引起其股票價格的變動。這這種情況下，他認為匯率的變動會引起股票價格的變化（傳統(tǒng)方法）。</p>

61、<p>　　Dornbusch和Boyer提出了一個表明股票價格與匯率相關(guān)的資本流動變動的模型。在股票價格下跌的時候，使得國內(nèi)的資本外流，降低了國內(nèi)貨幣的需求導致利率的下降，導致資本外流和貨幣貶值。</p><p>　　Bahmani-Oskooee和Sohrabian分析了美國國內(nèi)經(jīng)濟中的股價和匯率之間的關(guān)系。他們沒有得出在長期關(guān)系中兩者的相關(guān)關(guān)系，但他們使用了格蘭杰因果關(guān)系檢驗得出了短期內(nèi)兩者之間的

62、關(guān)系。Amihud，Bartov和Bodnar發(fā)現(xiàn)在不同時期，由于美元匯率的變化無法表明公司目前的股票收益率，所以兩者之間的關(guān)系是滯后的。</p><p>　　Ratner通過對1973年3月至1989年12月月度數(shù)據(jù)進行協(xié)整分析來測試美元匯率對美國股價的影響。他的研究結(jié)果表明，在長期中，美國股票價格與匯率之間是不存在因果關(guān)系的，因為即使是分時段的樣本都無法得到兩者之間的協(xié)整的關(guān)系。</p><

63、;p>　　Ajayi和Mougoue使用誤差修正模型分析了八個國家的股票價格與匯率之間的關(guān)系（加拿大，法國，德國，意大利、日本，荷蘭，英國和美國），他們發(fā)現(xiàn)這兩個變量之間顯著的短期和長期的關(guān)系。</p><p>　　Abdalla 和 Murinde使用從1985年1月至1994年7月期間的月度數(shù)據(jù)研究印度，韓國，巴基斯坦，菲律賓的四國的匯率與股價之間的關(guān)系。在一個誤差修正模型的框架中，他們證實了除菲律賓以

64、外的三個國家都存在股價與匯率之間的單向因果關(guān)系。他們通過格蘭杰因果關(guān)系檢驗驗證了匯率與股價之間的關(guān)系。</p><p>　　Ong 和 Izan使用1986年10月至1992年12月的澳大利亞和G - 7個國家的每周的“當前90天的點數(shù)”以及在澳大利亞，英國，法國和美國的“當前90天的點數(shù)”的期貨價格對兩者之間的關(guān)系進行研究。他們無法找到兩者之間的明顯的關(guān)系。他們猜想，使用日常的數(shù)據(jù)可能可以改善他們的試驗結(jié)果。&

65、lt;/p><p>　　Ajayi等人使用印度尼西亞和菲律賓的每日的數(shù)據(jù)對其股市和匯率之間的因果關(guān)系進行分析，同時認為韓國的匯率與股市之間的因果關(guān)系是與上述兩個國家相反的，而通過對香港，新加坡，泰國和馬來西亞的數(shù)據(jù)的觀察作者認為在這些國家兩者之間沒有因果關(guān)系，但，在臺灣，他們認為兩者之間存在雙向的因果關(guān)系。此外，同期在這八個國家中的只有三個國家兩者之間的關(guān)系有顯著的調(diào)整。而在發(fā)達國家中，他們發(fā)現(xiàn)貨幣市場和股票市場之間

66、有著顯著的單向因果關(guān)系。</p><p>　　Granger等人使用香港，馬來西亞，泰國和臺灣從1986年1月3日到1998年6月16日的每日的數(shù)據(jù)，發(fā)現(xiàn)匯率與股價指數(shù)之間有著顯著的雙向因果關(guān)系。此外，他們還發(fā)現(xiàn)對韓國的數(shù)據(jù)的研究的結(jié)果是與傳統(tǒng)的方法相一致的，但他們更加同意像菲律賓那樣的投資組合的方法。</p><p>　　Nieh和Lee使用恩格爾- Granger和Johansen的協(xié)

67、整檢驗發(fā)現(xiàn)在G - 7個國家中股票價格和匯率之間不存在長期的因果關(guān)系。此外，他們還發(fā)現(xiàn)這些國家在短期內(nèi)沒有確切的因果關(guān)系。然而，在一些國家中，無論是股票指數(shù)還是匯率都可以對一些變量進行預測。例如，他們發(fā)現(xiàn)在加拿大和英國貨幣的貶值會刺激股價在一天后上升。</p><p>　　實證研究表明，一般而言，大多數(shù)國家的這兩個金融變量（匯率與股票價格）之間是沒有長期均衡關(guān)系的。然而，許多的研究發(fā)現(xiàn)，雖然兩者之間的因果關(guān)系的方

68、向是根據(jù)該國具體特點所得出的，但這些變量具有“為對方進行預測的能力”。據(jù)我們所知，這篇文章是第一篇為巴西經(jīng)濟解決這個問題的論文。</p><p><b>　　數(shù)據(jù)和實證方法</b></p><p>　　我們的研究數(shù)據(jù)是使用1994年8月1日至2002年5月14日巴西證券交易所每日收盤價的股價指數(shù)和每日的匯率。因為如果我們使用的月度數(shù)據(jù)可能不足以發(fā)現(xiàn)其對資本流動的影響。

69、</p><p><b>　　1、單位根檢驗</b></p><p>　　我們用Augmented Dickey 和 Fuller（ADF）檢驗，一般來說，一個ADF（p）模型可表示為</p><p>　　ΔXt = α + (1 ? φ)Xt?1 + γt +ΣβiΔXt?i + εt.(3.1)</p><p>　　

70、用施瓦茨貝葉斯準則來選擇滯后變量（p）。如果得到的模型是與滯后數(shù)列序列相關(guān)的，那么除了滯后變量外，模型中還存在一個殘差項。</p><p>　　由于固定替代品的出現(xiàn)，使得Kwiatkowski等的單位根檢驗無法通過，因此，他們提出了通過讓空一個固定的替代品來作為單位根進行檢驗，這項測試是給出以下的方程：KPSS =1/T2ΣS2t/S2(L)，其中ST=Σei (t=1，2，3，……T)</p>