Loading...

Uncovered Interest Parity and Carry Trades

by Torsten Abendroth (Author)

Master's Thesis 2016 60 Pages

Business economics - Banking, Stock Exchanges, Insurance, Accounting

Excerpt

Table of contents

List of figures

List of tables

List of symbols

1 Introduction

2 Theoretical Framework
2.1 Uncovered Interest Parity
2.2 The Forward Premium Puzzle

3 Testing of UIP
3.1 Data
3.2 Regression equation and excess return formula
3.3 Results and analysis

4 Discussion of recent findings in literature
4.1 Approaches to explain a violation of UIP
4.1.1 Explanation by risk premium and peso problems
4.1.2 Explanation by irrationality
4.1.3 Explanation by learning problems and market inefficiency
4.2 Approaches to explain a partial holding of UIP
4.2.1 Developed and emerging markets
4.2.2 Short term and long term maturities

5 Consequences of recent innovations in the foreign exchange market

6 Conclusion

7 Bibliography

Appendix

List of figures

Figure 1: USD-AUD monthly log excess returns in % from 1984 to 2015

Figure 2: USD-AUD cumulative log excess return in % from 1984 to 2015

Figure 3: USD-NZD cumulative total, spot and carry returns from 1984 to 2015

Figure 4: USD-JPY cumulative log excess returns in % before and after costs

Figure 5: USD-AUD 1 Month ATM Volatility from 1995 to 2015

Figure 6: Average interest rate of the G10 3X3 Portfolio from 1989 to 2013

Figure 7: 1 - Month G10 Interest Rates from 1989 to 2013

List of tables

Table 1: OLS Regression Analysis

Table 2: Characteristics of Monthly Log Excess Returns in %

List of symbols

Abbildung in dieser Leseprobe nicht enthalten

1 Introduction

In research a large number of papers address the topics of Uncovered Interest Parity (UIP), the Covered Interest Parity (CIP) and consequently the carry trade as a popular investment strategy among investors. One way of executing a carry trade is by borrowing money in a low yielding currency and investing it in a high yielding currency. The range of assets in which can be invested is wide and reaches from equity and fixed income products to the target currency itself. Following Uncovered Interest Parity, a theory which is the basis for a lot of macroeconomic models, excess returns from carry trades are supposed to be zero on average. The underlying assumption is that the higher yielding currency should depreciate in a magnitude that offsets any gains from borrowing in the lower yielding currency. This is because rational and risk-averse market participants should expect inflation to rise in the target country. Yet empirical evidence shows that this assumption is not true and Uncovered Interest Parity is violated regularly. Since Eugene F. Fama (1984) struggled to satisfactorily explain his observation of high yielding currencies that do not depreciate but instead even gain in value, the phenomenon is known as the ‘forward premium puzzle’. The forward premium puzzle is the underlying reason why investment products like the Deutsche Bank Global Currency Harvest or the Barclay’s Capital Intelligent Carry Index ETN were engineered. They allow investors to invest in a carry trade basket with relatively low transaction costs and thereby to participate in a strategy which tends to outperform stock investments in terms of Sharpe-Ratios. Depending on the sample size and implemented methodology recent studies report Sharpe-Ratios which lie between 0.5 and 1.0 and thereby are exceeding Sharpe-Ratios close to 0.3 which are estimated for the equity market. (Norges Bank, 2014)

Nevertheless the sharp drawdowns of carry trades funded by the Japanese Yen during the Asia Crises in the late 1990s and during the Global Financial Crises in 2007 highlight that carry trades suffer in times of market distress. As will be shown in more detail this is due to an amplifying process which is triggered when the majority of investors simultaneously tries to unwind their often leveraged positions. Most recently the unexpected devaluation of the Renminbi (RMB) by the Chinese central bank led to market turmoil which in part was caused by market participants reducing or closing their long foreign exchange positions in the high yielding Chinese currency. Due to the tight trading band between the RMB and the US Dollar and therefore the low volatility environment, the risk-reward of investing in the high yielding RMB had been very attractive. The action of the Chinese authorities caught the investor community wrong footed which led to significant losses in other asset classes as well.

Over time different approaches to explain the excess return of carry trades emerged. They can be divided in an explanation by focusing either on a risk premium or a potential forecasting error resulting from irrational investors. On the other side a number of authors study peso and learning problems or market inefficiencies as a possible solution for the forward premium puzzle. Nevertheless there exists no single solution which researchers have agreed on. This is one of the main motivational reasons for writing this master thesis. In addition the availability of a higher quantity and better quality of data and the different underlying market environment compared to past studies are features which make a renewed look on a potential violation of UIP worthwhile.

The aim of this thesis is to test UIP by implementing an OLS regression analysis of five currency pairs which according to CFTC data, global turnover data and carry-to- risk ratios were among the most popular in the investor community. To increase the significance of this thesis for practitioners, the work will use one month forward contracts which are used frequently by investors and include bid and ask rates in order to account for transaction costs. In addition all currency pairs include the US Dollar due to reasons for better liquidity and therefore tighter bid-ask spreads. Moreover the work will present recent findings in literature which try to explain deviations from UIP. Approaches can be separated by the focus on a risk premium, on irrational market behavior or on learning problems and market inefficiency. While most focus laid on an explanation by a risk premium it will be shown that it is crucial to combine the different scientific disciplines in order to solve the forward premium puzzle.In addition the thesis will provide an outlook on the future attractiveness of carry trade strategies. This projection will be given by using combination of the knowledge which was accumulated throughout the thesis and recent innovations in the foreign exchange market. In order to achieve these goals the thesis will start with an overview of the underlying theoretical framework. This will be done by presenting the relationship between UIP, CIP and Real Interest Rate Parity and a presentation of the early findings of Fama. Afterwards the underlying data and the regression and excess return formulas will be described. This is followed by a presentation of the results from the regression analysis and a discussion of the log excess return characteristics. In the next step existing explanations of deviations from UIP will be presented, classified and discussed in order to look for weaknesses which have to be addressed in future models. After an outlook on the future attractiveness of carry trades in the next chapter, the last chapter will conclude the most important findings of this thesis.

2 Theoretical Framework

In order to explain the significance of the Uncovered Interest Parity it makes sense to put theory in a broader macroeconomic framework. This facilitates an interpretation of the results and a judgement of the different existing solutions in literature. Afterwards the way Fama (1984) performs his OLS regression analysis will be described and some of his findings presented.

2.1 Uncovered Interest Parity

Uncovered Interest Parity describes the relationship between domestic and foreign nominal interest rates and the expected change of the current spot exchange rate and can be stated as follows:

Abbildung in dieser Leseprobe nicht enthalten

In this equation [Abbildung in dieser Leseprobe nicht enthalten] represents the expected future spot exchange rate in +1 determined by the market at time . Thereby equation (1) implies that [Abbildung in dieser Leseprobe nicht enthalten] which according to Lewis (1995) is the ‘statistical expectations operator conditional on time information’ equals ‘the market’s expectation conditional upon current information’ [Abbildung in dieser Leseprobe nicht enthalten] (p. 1917). The statistical expectations should equal the market expectations assuming rational investors but as will be pointed out later some authors try to explain deviations from UIP by forecasting errors and irrationality which can lead to [Abbildung in dieser Leseprobe nicht enthalten] stands for the current spot exchange rate. The nominal interest rates are expressed by [Abbildung in dieser Leseprobe nicht enthalten]for the domestic nominal interest rate which in this work is always the US interest rate and by [Abbildung in dieser Leseprobe nicht enthalten] for the return on a comparable foreign asset. According to equation (1) the market determined future spot exchange rate should exactly offset the profit a rational and risk-neutral investor can generate in efficient markets from the interest differential between two currencies. The underlying economic assumption is the international Fisher effect which states that the high yielding country should experience higher inflation pressures and consequently a depreciation of its currency.

UIP has its name due to the absence of tradable forward contracts and does not resemble a no-arbitrage condition. In contrast the CIP describes a no-arbitrage condition in a market where forward contracts are available to cover the foreign exchange exposure and is written as follows:

Abbildung in dieser Leseprobe nicht enthalten

The theory holds that when the forward rate is priced as such there exists no risk free profit opportunity at time t. This is the case when an investor has no gains from borrowing a foreign currency at [Abbildung in dieser Leseprobe nicht enthalten]changing the amount in US Dollar at the current spot rate and selling the proceeds forward at the current forward rate . The arbitrage-free forward pricing in the CIP theory is linked to the Law of one Price (LOP) which states that in an efficient financial market identical securities should have the same price independently of the way they are created or in which market they are traded. For a forward contract this implicitly means that the price at which it can be traded should not deviate from the price which can synthetically be reproduced by using loans and deposits or bonds denominated in two different currencies.

A lot of papers analyzed empirical evidence of deviations from CIP by using market data. The results are important because a confirmation of a holding of CIP allows for testing the UIP theory by using forward exchange rate data. Akram, Rime and Sarno (2008) provide an innovative approach to test deviations from CIP by using high frequency tick data. They conclude that short-lived violations of CIP can arise and the violations can be economically significant. Moreover the deviations are high enough to be exploitable for agents but their duration is low enough to suggest that market exploits possible arbitrage opportunities quickly. Additionally the chances for arbitrage are very unpredictable, which leads them to the conclusion that in research one can assume CIP to hold (p. 251). Furthermore Taylor (1989) finds no evidence of exploitable profit opportunities in a calm market environment but small and exploitable arbitrage possibility in times of market turbulence. For the latter periods he reports few but hardly any evidence of arbitrage opportunities in the short term but influential and sometimes consistent arbitrage chances in longer maturities. Nevertheless Taylor summarizes that the deviations of CIP can be linked to liquidity constraints and credit limits and therefore supports the covered interest rate parity theorem as well (p. 389). In a similar way Fong, Valente, Fung (2010) argue that deviations from CIP can be seen as a compensation for credit and liquidity risk. For the US Dollar - Hong Kong Dollar market they find evidence of CIP deviations and are able to show that after adjusting for the two risk components and transaction costs there tends to be no profit left (p. 1106).

As a consequence the work will follow the suggestion of the citied findings and assume CIP to hold in order to test Uncovered Interest Parity. Given that CIP holds one can combine Equations (1) and (2) to get:

Abbildung in dieser Leseprobe nicht enthalten

This relation is called the unbiased forward hypothesis and is true when both UIP and CIP hold. It basically means that the forward rate is an unbiased predictor of the expected future spot exchange rate. Therefore forward contracts can be used to test whether this holds empirically in financial markets. The formula will be adjusted by using logs and subtracting the current spot exchange on both sides of the equation to get our regression equation in the following section.

An important aspect to note is that there seems to exist ’some confusion in the literature on the relation of uncovered interest parity to purchasing power parity and real interest rate equality’ (Engel, 1996, p. 136). In order to show the true relationship between the three components Engel (1996) decomposes deviations from rational expectations real interest rate parity (DRRIP) into the sum of rational expectations uncovered interest parity (DRUIP) and deviations from ex ante purchasing power parity (DEPPP) . While goods market integration implies ex ante purchasing power, capital market efficiency together with risk neutrality implies rational expectations uncovered interest rate parity. Thus he criticizes the approach of explaining deviations from UIP by separating between failures of PPP and real interest rate parity because real interest Parity itself ’has no meaning independent of the notion’ (p. 136). Most importantly Engel (1996) reported that failure of UIP can be explained not only by non-rational expectations but also by a risk premium in case the investors are not risk neutral and do not have rational expectations (p. 138). In the following part the topic of a potential risk premium as a possibility to explain deviations from UIP will be addressed by presenting the approach of Fama who divides the forward rate in a rational forecasting component and a premium.

2.2 The Forward Premium Puzzle

Hodrick (1980) and Fama (1984) were one of the first authors to discuss a risk premium component in the determination process of the forward exchange rate. Hodrick (1980) rejects the simple efficient-market hypothesis by not rejecting the rationality of foreign exchange markets itself but instead by adding the assumption of risk-averse behavior of market participants. He concludes that his findings suggest that fluctuation in the risk premium could be one source of explanation (p. 831). Indeed under the assumption of efficient and rational markets Fama (1984) finds evidence of a time-varying risk premium. He comes to this conclusion by implementing the following two OLS regression equations:

Abbildung in dieser Leseprobe nicht enthalten

In equation [Abbildung in dieser Leseprobe nicht enthalten] represents the premium plus the random error of rational forecast [Abbildung in dieser Leseprobe nicht enthalten] which implies that if [Abbildung in dieser Leseprobe nicht enthalten] is significant different from zero the premium component of [Abbildung in dieser Leseprobe nicht enthalten] shows variation that can be observed in െ +1. Equation (5) examines whether the forward-spot differential at time t has power to predict the difference between the future spot rate [Abbildung in dieser Leseprobe nicht enthalten] and [Abbildung in dieser Leseprobe nicht enthalten] This is the case if 2 is reliably non-zero (p. 320). Fama uses the two regression coefficients resulting from the two stated equations in order to derive the following relationship.

Abbildung in dieser Leseprobe nicht enthalten

Using this relationship Fama shows that the variance of the risk premium component is higher than the variance of the expected future spot rate (p. 337). Even though Fama implements two regression equations he notes the most used approach was to analyze only the results of equation (5) and to test if 2 is significantly different from one (p. 321) For his findings of even negative slope factors for regression (5) and the high variance of the future change in the spot rate compared to the ex ante forward- spot differential Fama has no convincing and satisfying answer. Fama (1984) discusses existing explanations which for example aimed at solving the forward premium puzzle by using forecasting errors as one source of explanation. Another approach was to examine government interventions which for instance resulted in the expansion of traditional models like CAPM or APT by adding a monetary policy factor. Moreover doomsday theories which resulted in the study of peso problems and downside risk and the analysis of stochastic deviations from purchasing power parity as another approach were studied as well (pp. 334-336). Some of the early approaches will be picked up later and it will be shown how the different paths of explanations developed and which new approaches were added over time. To summarize chapter 2 it can be said that UIP plays an important role in macroeconomic models and that CIP needs to hold in order to test for deviations from UIP by using forward contracts. In addition the underlying assumption of risk- neutral and rational investors may not hold in practice which leads to possible explanations by a risk premium or irrational market expectations. Fama was one of the first authors to identify a time-varying risk premium as an explanation of the forward premium puzzle.

In order to compare the results of this study to those of Fama and to findings in literature, UIP will be tested by using recent data for bilateral currency pairs. In the next chapter the starting point will be a description and presentation of the underlying data set. In the second step the regression results and excess returns characteristics of our exchange rate data will be presented and analyzed.

3 Testing of UIP

While in recent literature the majority of papers examine carry trade excess returns by using either baskets of G10 currencies or a combination of developed markets and emerging markets currencies this thesis will focus on bilateral currency pairs instead. This is because in practice a huge part of carry trades is implemented with bilateral currency pairs. Brunnermeier, Nagel and Pedersen (2009) for example report that investors directly use specific currency pairs like the Australian Dollar against the Japanese Yen and exclude the US Dollar as base currency in order to profit from the high yield spread between the AUD and the JPY (p. 320). Nevertheless it is important to note that bilateral currency pairs are characterized by substantial noisiness as explained in Atanasov and Nitschka (2014, p. 274).

Explaining excess returns by using large baskets or portfolios of currencies means that idiosyncratic or country-specific risks do not need to be explained anymore and researchers can focus on explaining the systemic risk components. Our data set of bilateral currencies therefore is exposed to two types of risk components which will be taken into account in chapter 4.

3.1 Data

The data set consists of end-of-month spot and forward rates which are collected from Barclays Bank International and Reuters via Reuters Datastream. In addition Bloomberg Data for the presentation of current and historical 1 month at-the-money option volatilities will be used. The analysis includes the currency pairs US Dollar (USD) against Swiss Franc (CHF), US Dollar against New Zealand Dollar (NZD), US Dollar against Japanese Yen (JPY), US Dollar against Australian Dollar (AUD) and US Dollar against Mexican Peso (MXN). The US Dollar is used in every currency pair due to reasons of availability and quality of data. Moreover currency pairs quoted against the US Dollar tend to have tighter bid-ask spreads and therefore higher liquidity. The lower transaction costs due to the availability of narrow bid-ask spreads lead to a higher attractiveness among investors and make it easier for traditional models, which rely on efficient markets without transaction costs, to explain the existence of excess returns.

The length of the time series and therefore the number of sub periods differ between the currency pairs, with the longest time series for the Swiss Franc and Japanese Yen ranging from October 1983 to September 2015 and therefore including 383 sub periods. The exchange rates are defined as units of foreign currency per US Dollar and following Lustig, Roussanov and Verdelhan (2014) mid prices are used for the regression analysis and bid and offer prices are used for calculating excess returns after transaction costs. The reason for choosing these five specific currency pairs is the evidence one can extract from CFTC Data, carry-to-risk ratios and of market turnover data, which all indicate that the currencies belonged to the most popular currencies among investors. Curcuru, Vega and Hoeck (2010) show that even though CFTC Data may be misinterpreted because for instance some big market participants like hedge funds trade mainly OTC, the net short positions in USD-CHF Futures peaked in 2006-2007 at a time when media accounts also showed that the Swiss Franc was a very popular funding currency (p.6). Moreover CFTC Positioning shows that CHF, JPY were amongst the most shorted currencies between 1983-2013 while the AUD, NZD, MXN were amongst the most bought currencies (Norges Bank, 2014, p. 26). Moreover the carry-to-risk ratio which sets the interest rate differential of two currencies in relation to their volatility implied by ATM options can serve as a measure of attractiveness. A higher carry-to-risk ratio implies higher popularity. The positive correlation between futures positioning and carry-to-risk ratio for the AUD and the negative correlation for the CHF and the JPY increases the validity of the two indicators. (Galati, Heath, & McGuire, 2007, p. 38f)

In addition the quality and quantity of data for the sample G10 currencies and the Mexican Peso is superior to a number of other currencies for instance in terms of not violating Covered Interest Rate Parity or pegged currency regimes1, no missing data for bid-offer spreads, a longer time series and no implausible data.

3.2 Regression equation and excess return formula

Given that Covered Interest Parity holds one can test the Unbiased Hypothesis by combining Equations (1) and (2), using logs and adding an error term to get regression equation:

Abbildung in dieser Leseprobe nicht enthalten

The regression in form of (6) basically equals Fama’ regression equation (5) and is the most used one scientific literature to test for violation of UIP (Flood & Rose, 2002, p. 253). [Abbildung in dieser Leseprobe nicht enthalten] is the log of the observed spot rate in [Abbildung in dieser Leseprobe nicht enthalten] while is the log of current spot rate. The difference [Abbildung in dieser Leseprobe nicht enthalten] represents the change in the spot rate. The intercept term is described as [Abbildung in dieser Leseprobe nicht enthalten], the regression coefficient as [Abbildung in dieser Leseprobe nicht enthalten] and the random error as [Abbildung in dieser Leseprobe nicht enthalten]. Lastly[Abbildung in dieser Leseprobe nicht enthalten] represents the log forward discount consisting of the log forward exchange rate at time t and the log spot exchange rate at time t. To derive the regression equations (4) and (5), Fama (1984) separates the forward rate in a risk premium and an expected future spot rate, to express the “market determined certainty equivalent of the future spot exchange rate" (p.320) in the mathematical form of:

Abbildung in dieser Leseprobe nicht enthalten

Another way to mathematically include the risk premium component will be presented in chapter 4. In the next step he subtracts the current spot exchange rate on both sides of the equation to generate his two regression equations. He uses logs for both the spot and the forward rates and covers the time frame from August 1973 to December 1982. Additionally, Fama (1984) finds higher autocorrelation for the forward discount, Ft-St, compared to low autocorrelation for the change in the spot rate, St+1-St. He explains that the autocorrelation of [Abbildung in dieser Leseprobe nicht enthalten] is buried by the high variability of the unexpected components [Abbildung in dieser Leseprobe nicht enthalten] and [Abbildung in dieser Leseprobe nicht enthalten] (p. 325). In the results of this study Fama’s findings regarding the higher standard deviation of [Abbildung in dieser Leseprobe nicht enthalten]compared to the current log forward discount can be discovered as well.

The Null Hypothesis which will be tested by using regression equation (6) can be written as follows:

Abbildung in dieser Leseprobe nicht enthalten

If the Null Hypothesis can be rejected with a two tailed test at a significance level of 5% there would be evidence for the violation of UIP. While some papers are also testing the Null Hypothesis that 1 is equal to 0 and in some cases perform a joint test for the two described conditions this work only tests if the slope coefficient equals unity. Moreover this work is not using overlapping data which for example would be more appropriate for currency pairs with a short available time series or for testing UIP with longer maturity bonds.

To get a deeper understanding of the characteristics of the sample currency pairs their monthly log excess returns will be described by calculating their mean, standard deviation, skewness and excess kurtosis. In order to show the overall performance of the different carry trades and especially their sharp drawdowns in times of market distress excess returns will be calculated as well. In addition it will be shown how much of the currency specific cumulative return is recruited from the pure spot exchange rate movement and how much stems from the carried interest. From the point of view of an US investor the excess return formulas for buying or selling a foreign currency in the forward markets excluding transaction costs are as follows:

Abbildung in dieser Leseprobe nicht enthalten

According to Atanasov (2014) the advantage of using forward contracts instead of using interest rate differentials is that one can capture the effect of including transaction costs by using bid-ask spreads (p. 273). Following Fama (1984) the excess returns as well as the differences in the regression analysis are multiplied with 100 in order to get percentage numbers. The excess returns after transaction costs will be calculated by using (13) for an US Investors who is long the foreign currency in the forward market and (14) for a US person who is short the foreign currency in the forward market.

Abbildung in dieser Leseprobe nicht enthalten

While the term represents the bid rate at which the US investor can sell US Dollar the term stands for the ask rate at which the US investors can buy US Dollar and sell the foreign currency.

In the next sub chapter the results for the regression analysis and the excess returns will be shown. The way of presenting our numbers will be similar to the one Fama used in order to make a comparison between both findings easier.

3.3 Results and analysis

In this sub chapter the focus will lie on interpreting our calculation results. The following table summarizes our OLS Regression results. In addition the statistical significance and the standard deviations values for the log forward discount and the change in the log spot exchange rates are presented.

Table 1: OLS Regression Analysis

Abbildung in dieser Leseprobe nicht enthalten

Reference: Datasource - Barclays Bank International via Datastream

The underlying OLS Regression is as follows:[Abbildung in dieser Leseprobe nicht enthalten]

α1 is the Intercept, β1 is the regression coefficient, while s(α1) and s(β1) are the standard errors of the estimated regression coefficients. R2 is the coefficient of determination and s(ε1) represents the residual standard error. DF stands for Degrees of Freedom, T-Stat is the t-statistic for a two tailed test of the Null Hypothesis in the form of Ho: β1 = 1 at a significance level of 0.05 and P is the resulting p-value. The expressions σft[Abbildung in dieser Leseprobe nicht enthalten] represent the standard deviations of the log forward discount and the observed change in the spot rate. The main result is that the Null Hypothesis for all our currencies can be rejected and UIP does not hold for this sample. The regression coefficients are all significantly different from 1 at the 5% level and only the p-value of the Swiss Franc against the US Dollar is close to the 5% rejection level. Moreover the coefficients of determination are all close to zero which is due to high standard deviation of the change in the spot rate compared to the standard deviation of the log forward discount [Abbildung in dieser Leseprobe nicht enthalten] what confirms the finding of Fama. For all currency pairs negative values for the coefficient of determination can be found which reaffirms that the forward premium puzzle is still at work. The standard errors and coefficients of determination are in line with the findings of Fama even though he uses a smaller sample of data and did his research in a time when most currencies were not floating as freely as in later periods. Furthermore the very negative 1values for the US Dollar long position against the Japanese Yen and against the New Zealand Dollar indicate that the cross exchange rate between New Zealand and Japan delivered significant returns as well as can be seen in Gyntelberg and Remolona (2007, p. 76).

Reasons for the strong evidence for a violation of UIP could be the public availability of CFTC positioning data, carry-to-risk ratios and the inclusion of the currencies in tradeable carry trades baskets like the Deutsche Bank Global Currency Harvest. Providing information on the speculative positions and the attractiveness of carry trades and enabling investors to invest in a portfolio of carry trades at low costs may lead to additional inflows which are pushing the slope coefficient further away from unity. Trying to solve the forward premium puzzle by arguing with directional, exaggerated and irrational inflows due to data on the positioning of other investors or carry-to-risk ratios would be classified as an explanation by irrationality. An explanation of a failure of UIP connected with low transaction costs mainly addresses the topic of market frictions and market inefficiencies. Yet approaching the violation of UIP with irrationality or market inefficiencies is not as popular as using a risk premium. In order to get a deeper understanding of the characteristics of the data Table 2 aggregates the statistical moments on a monthly basis and shows which part of the total cumulative log excess return is due to spot movement and which part due to earning the interest rate differential.

[...]


1 Lustig et al. (2014) excluded observations from Hong Kong, Saudi Arabia and United Arab Emirates due to pegged currency regimes and for instance excluded the South African Rand (ZAR) for a short period in 1985 due to large failure of CIP.

Details

Pages
60
Year
2016
ISBN (eBook)
9783668382114
ISBN (Book)
9783668382121
File size
1 MB
Language
English
Catalog Number
v351449
Institution / College
University of Frankfurt (Main) – Goethe Business School
Grade
1,0
Tags
Carry Trade Uncovered Interest Rate Parity UIP Covered Interest Rate Parity CIP Uncovered Interest Parity Covered Interest Parity Forward Premium Puzzle OLS Regression Analyis Peso Problem Risk Premium Purchasing Power Parity TED Spread VIX Learning Problem

Author

  • Depending on your browser, you might see this message in place of the failed image.

    Torsten Abendroth (Author)

Previous

Title: Uncovered Interest Parity and Carry Trades