本篇論文主要在外匯市場建立市場投資組合、利差交易投資組合與動能交易投資組合，探討透過不同情境適當改變投資組合比重配置，是否能夠顯著提升交易策略的報酬表現。 以1999年1月至2015年10月為樣本期間，根據28個國家外匯市場資料建構市場投資組合、利差交易投資組合與動能交易投資組合等，之後根據三種投資組合報酬情況透過馬可夫情境轉換模型區分成三種情境。按三種情境的各種投資組合超額報酬表現，再利用馬可維茲的平均數-變異數投資組合模型配置各情境下各項交易的比重，再依據計算出的預期情境與相對應比重進行投資。其結果顯示在樣本期間內，本篇論文的交易策略相較於外匯市場投資組合、利差交易投資組合與動能交易組合有較佳的投資表現。 在樣本外測試部分，採用自2012年中開始的連續情境二資料進行分析。報酬方面，在其他交易型態呈現負報酬較多情況下，就本文交易策略而言，投資者隨時根據其各種交易平均報酬與共變異數進行交易比重配置，適時放空交易策略或投資無風險資產，產生正報酬。但從標準差可以推斷投資者面對未來的不確定，在整個樣本外期間歷時的34個月當中標準差亦無法有效降低，說明了投資者面對下一期總體環境的高不確定性。 / In this thesis, we mainly investigate whether it could improve the performance of currency portfolio by adjusting weights among carry trade, momentum and market return in foreign exchange market under different kinds of regimes. Based on a sample of 28 market currencies, we form three kinds of transactions in our portfolio, including carry trade, momentum, and market return. Under Markov switching model, we divide the sample period into three regimes, and then determine weights among carry trade, momentum and market return by parameters of each re-gime using Markowitz mean-variance analysis. Finally, we invest different weights among three transactions according to each expected regime. We find the result that although the return of the strategy is just a little higher than the carry trade, the risk is much lower compared to other transactions. In our out-of-sample testing, we analyze the performance by using the data of the regime two which begins September, 2012. With the respect to the return, most of other risky transactions have negative return, but we get positive return by adjusting the long position and short position according to the result of the mean-variance anal-ysis. However, we can not effectively reduce risk by using the strategy, and in the meantime it can explain the high uncertainty investors face toward the next period.

利差交易(Carry Trade)是在各種貨幣的利率水平上進行套利的交易以獲得價差收益的交易型態。而利差交易在這幾年創造出諸多研究的方向；利差交易又稱為套息交易，即借入低利率的貨幣，以購買其它高收益的投資工具，如高利率貨幣、股票、或是實物資產等，進而從中賺取其間的利率及匯率差價。諸如債券或國庫券等現貨金融工具所帶來的收益與該項投資的融資成本間的差額。 利差交易多在市場處於穩定低風險狀況下時，才能夠得到穩定的報酬率，因此如何客觀準確的評量目前市場所處的風險狀態，以獲得相對平穩又較佳的收益是主要的研究方向。本研究以如何求得利差交易裡面各項最佳且保持穩定性的參數與指標，假設利差交易可以經由衡量某些風險指標的平均值，並當風險指標低時建立利差交易部位，反之，當風險指標高時結束利差交易，甚至更積極進行反向利差交易，以求達到穩定報酬率的目標。 / Title of Thesis: Empirical Performance of Carry Trade Trading Strategy School/Graduate School: National Chengchi University Executive Master Of Business Administration, Advanced Finance Class – Risk Management and Insurance Group Graduate Student : Lee, Nai-Chun Instructor: Dr. Kang, Jung Pao Thesis Content: Carry trade bases on interest rate differences of many currency pairs to make capital gains and interest income and it creates many research topics in recent years. Carry trade borrows low yield currencies and invests other high yield targets such as high yield currencies, stocks, or real assets to get interest incomes and capital gains from foreign exchange rates. Only when a foreign exchange market is under stable and low-risk conditions, carry trade can achieve stable return rates. Therefore, how to objectively evaluate current market risk situations to get relatively stable and more returns is the main research topic of this thesis. This thesis reports how to get optimal and consistently stable parameters and indicators of carry trade. It assumes that carry trade can build positions by measuring some mean values of risk indicators when risk indicators are low. On the contrary, it ends the trade when risk indicators are high and even actively short carry trade positions to achieve stable return rates. Key words: carry trade、capital gain、exchange rate、risk indicator

One of the most popular investment and trading strategies over the last decade, has been the currency carry trade, which allows traders and investors to buy high-yielding currencies in the Foreign Exchange spot market by borrowing, low or zero interest rate currencies in the form of pairs, such as the Australian Dollar/Japanese Yen (AUD/JPY), with the purpose of investing the proceeds afterwards into fixed-income securities.To be able to determine the causality between the returns of equity markets and the foreign exchange market, we choose to observe the sensitivity and influence of two equity indexes on several pairs involved in carry trading. The reason for studying these relationships is to further explain the causes of the uncovered interest parity puzzle, thus adding our contribution to the academic field through this thesis.To accomplish our goals, data was gathered for daily quotes of 16 different currency pairs, grouped by interest differentials, and two equity indexes, the S&P 500 and FTSE All-World, along with data for the VIX volatility index, for the 2003-2012 period. The data was collected from Thomson Reuters Datastream and the selected ten year span was divided into three different periods. This was done in order to discover the differences on how equity indexes relate to typical carry trade currency pairs, depending on market developments before, during and after the world financial crisis.The tests conducted on the collected data measured the correlations, influences and sensitivity for the 16 different currency pairs with the S&P 500 Index, the FTSE All-World index, and the volatility index between the years of 2003-2012. For influences and sensitivity, we performed Maximum Likelihood (ML) regressions with Generalized Autoregressive Conditional Heteroscedasticity (GARCH) [1,1], in Eviews software.After analyzing the results, we found that, during our chosen time period, the majority of currency pair daily returns are positively correlated with the equity indexes and that the FX pairs show greater correlation with the FTSE All-World, than with the S&P 500. Factors such as the interest rate of a currency and the choice of funding currency played an important role in the foreign exchange markets, during the ten year time span, for every yield group of FX pairs.Regarding the influence and sensitivity between currency pairs and the S&P 500 with its VIX index, we found that our models explanatory power seems to be stronger when the interest rate differential between the currency pairs is smaller. Our regression analysis also uncovered that the characteristics of an individual currency can show noticeable effects for the relationship between its pair and the two indexes.

Submitted by Alexis Petri Magalhaes Costa (alexis.petri.costa@gmail.com) on 2017-02-22T19:36:54Z No. of bitstreams: 1 Dissertacao.pdf: 1279843 bytes, checksum: b21e4e76b5b7742d95c444d256cd3ec9 (MD5) / Rejected by Renata de Souza Nascimento (renata.souza@fgv.br), reason: Alexis, boa tarde Para que possamos aceitar seu trabalho, por gentileza, retire as páginas em branco que constam no trabalho entre a capa, contracapa e folha de assinaturas. Inclua também o resumo em português (mesmo o trabalho estando em inglês). Após correções, submeter o arquivo novamente. Att on 2017-02-22T19:41:54Z (GMT) / Submitted by Alexis Petri Magalhaes Costa (alexis.petri.costa@gmail.com) on 2017-02-23T15:29:04Z No. of bitstreams: 1 Dissertacao.pdf: 1281154 bytes, checksum: 674d2d72b6ef337a490c5cc1124c144e (MD5) / Approved for entry into archive by Renata de Souza Nascimento (renata.souza@fgv.br) on 2017-02-23T15:51:19Z (GMT) No. of bitstreams: 1 Dissertacao.pdf: 1281154 bytes, checksum: 674d2d72b6ef337a490c5cc1124c144e (MD5) / Made available in DSpace on 2017-02-23T17:25:05Z (GMT). No. of bitstreams: 1 Dissertacao.pdf: 1281154 bytes, checksum: 674d2d72b6ef337a490c5cc1124c144e (MD5) Previous issue date: 2017-01-31 / This work assesses the validity of applying the Taylor Rule to the Brazilian market. Several variables, tools and features were analyzed. Among its variables, inflation, inflation target and output gap were included to determine a basis scenario. On top of that, exchange rates and exchange rate related information was tested. Both the crude market input (spot rate) and a trade-weighted currency are included in this analysis. Also extracted from the market, the carry-trade premium was calculated from future exchange rate quotes. Among its tools, the smoothing factor was evaluated. The series were tested for regime breaks, unit root and cointegration, residual autocorrelation, normality, heteroskedasticity and coefficient linearity. Among its features, special attention was paid to the proper timing of each variable. The regressions being forward looking, it was important to line up the actual information available for the Brazilian Central Bank at the time of each decision. Timing was again a factor when considering different cutoff periods, and for synchronizing market data, especially for constructing the carry-trade payoff. This work concludes that evidence of a Taylor Rule being a response function for the Brazilian Central Bank is shaky, especially given the number of misspecification indicators found. Results also suggest that, assuming there is a need to protect the local economy from sharp capital flows consequent of interest rate changes, the implicit future exchange rate premia is not a good indicator of such risk. / Esse trabalho avalia a aderência da Regra de Taylor à política monetária brasileira. Diferentes variaveis, ferramentas e características foram avaliadas. Entre suas variáveis, essa dissertação avaliou inflação, gap de inflação, e hiato do produto, para determinar um cenario base. Sobre este, informações referentes a câmbio foram testadas. Fez-se regressões incluindo o spot de mercado e o câmbio ponderado pela balança comercial. Testou-se também o prêmio implícito em um carry-trade hipotético utilizando o primeiro futuro de dolar da BM&F. Entre as ferramentas estudadas, o 'smoothing factor' foi analisado e não foram encontradas melhorias significativas. As séries foram testadas para quebras de regime, raiz unitária, cointegração, autocorrelação dos resíduos, normalidade, heteroskedasticity e linearidade de coeficientes. Entre suas características, esta dissertação leva em consideração o 'timing' de cada variável. As regressões sendo 'forward looking', buscou-se exatamente o valor para cada variável disponível ao Banco Central do Brasil no momento de cada decisão do COPOM. Esse mesmo cuidado foi tomado para sincronizar os dados de mercado, especialmente para construir o 'payoff' do carry-trade. Esta dissertação conclui que há evidências apenas fracas de que a função resposta do Banco Central esteja em linha com a Regra de Taylor, especialmente dado o número de problemas de especificação encontrados. Os resultados também sugere que, supondo que haja a inteção de proteger a economia local de choques de fluxo de capital consequentes de mudança na taxa SELIC, o prêmio implícito no 'carry-trade' não é um bom indicador desse risco.

Submitted by Pablo F. Oliveira (pablo.perque@gmail.com) on 2012-02-29T12:36:59Z No. of bitstreams: 1 Dissertacao - Pablo Frisanco Oliveira -final.pdf: 1039827 bytes, checksum: 63d8e3ff3c6593d9ef449829e78e77c1 (MD5) / Approved for entry into archive by Gisele Isaura Hannickel (gisele.hannickel@fgv.br) on 2012-02-29T12:45:22Z (GMT) No. of bitstreams: 1 Dissertacao - Pablo Frisanco Oliveira -final.pdf: 1039827 bytes, checksum: 63d8e3ff3c6593d9ef449829e78e77c1 (MD5) / Made available in DSpace on 2012-02-29T12:55:43Z (GMT). No. of bitstreams: 1 Dissertacao - Pablo Frisanco Oliveira -final.pdf: 1039827 bytes, checksum: 63d8e3ff3c6593d9ef449829e78e77c1 (MD5) Previous issue date: 2012-01-30 / The aim of the present work is verify if, when the higher moments (skewness and kurtosis) are taken in consideration for carry trade portfolio allocation optimization, an investor can be better off than the traditional allocation, which prioritizes only the first two moments (mean and variance). The hypothesis of the research is that a carry trade currency exhibits non-Normal returns distribution, and its higher moments have a dynamic which can be modeled by GARCH-type model, in this specific case IC-GARCHSK. This model consists of one equation to each of the independent components’ conditional moments, named the returns, variance, the skewness, and the kurtosis. Another hypothesis is that a CARA (constant absolute risk aversion) utility function investor can have its function approximated by 4th order Taylor expansion. The work’s strategy is modelling the dynamics of the daily log-returns series’ moments of some carry trade currencies using the model above and dynamically estimate the optimal allocation which maximizes the investor’s expected utility function. The results show that the investor can benefit from taking in consideration the series’ higher moments, once this portfolio exhibited smaller opportunity cost than one that uses only mean and variance as criteria. / O objetivo do presente trabalho é verificar se, ao levar-se em consideração momentos de ordem superior (assimetria e curtose) na alocação de uma carteira de carry trade, há ganhos em relação à alocação tradicional que prioriza somente os dois primeiros momentos (média e variância). A hipótese da pesquisa é que moedas de carry trade apresentam retornos com distribuição não-Normal, e os momentos de ordem superior desta têm uma dinâmica, a qual pode ser modelada através de um modelo da família GARCH, neste caso IC-GARCHSK. Este modelo consiste em uma equação para cada momento condicional dos componentes independentes, explicitamente: o retorno, a variância, a assimetria, e a curtose. Outra hipótese é que um investidor com uma função utilidade do tipo CARA (constant absolute risk aversion), pode tê-la aproximada por uma expansão de Taylor de 4ª ordem. A estratégia do trabalho é modelar a dinâmica dos momentos da série dos logartimos neperianos dos retornos diários de algumas moedas de carry trade através do modelo IC-GARCHSK, e estimar a alocação ótima da carteira dinamicamente, de tal forma que se maximize a função utilidade do investidor. Os resultados mostram que há ganhos sim, ao levar-se em consideração os momentos de ordem superior, uma vez que o custo de oportunidade desta foi menor que o de uma carteira construída somente utilizando como critérios média e variância.

本研究對1985/1至2016/10期間，37種貨幣的超額報酬與國家風險進行實證分析，以The PRS Group發佈的ICRG綜合風險評級做為國家風險的衡量指標。各國貨幣分別進行時間序列分析的結果顯示，單一國家的國家風險與該國貨幣的匯率走勢及超額報酬並不存在顯著的關聯。 投資組合分析的結果，對高國家風險貨幣與低國家風險貨幣分別執行利差交易，結果顯示兩者的超額報酬並沒有顯著差異。而動能策略在高國家風險貨幣則可以獲得顯著較高的超額報酬。 Fama-Macbeth二步驟迴歸分析結果顯示，高國家風險的投資組合確實擁有較高的因子負載量，然而國家風險因子的市場價格，也就是承受一單位 β_CRISK獲得的國家風險溢酬太低不顯著，因此國家風險無法幫助解釋貨幣報酬。 / We empirically investigate the relation between currency excess returns and country risk, as measured by the ICRG comprehensive risk rating issued by The PRS Group, of 37 currencies during 1985/1 to 2016/10. The result of the single currency time series analysis shows that there is no significant correlation between the country risk and the exchange rate movement, also the currency excess return. As a result of the portfolio analysis, there is no significant difference in excess returns when we execute carry trade respectively on high country risk currencies and low country risk currencies. While the momentum strategy in the high country risk currencies can generate significantly higher excess return. The results of the Fama-Macbeth two-step regression show that the high-risk portfolios do have a higher factor loading, whereas the country risk factor's market price, that is, the country risk premium received by a unit of β_CRISK, is too low. Therefore, country risk cannot help explain currency excess return.

本研究主要是以「外匯報酬三因子模型」為基礎，故先檢視在本樣本期間內(1985/2至2016/10) ，以雙分類法將37國主流貨幣分為9個投組後，外匯超額報酬解釋力，是否會因加入動能策略因子形成之三因子模型，而較原本兩因子模型(市場因子、利差策略因子)來的強？最終測得三因子模型在判斷係數及殘差等適切度表現較佳。 接著利用逐步迴歸分析法(限制所有自變數均須於90%信心水準內顯著)嘗試尋找獲利成因，主要挑選出不同面向之11種經濟成因因子(股價指數波動、投機活動、流動性、貨幣波動、落後短期利率、落後股利率、落後期限利差、落後違約利差、)落後避險基金套利資本、工業生產量及通膨率因子)來檢測可否解釋三因子模型中獲取報酬之利差、動能策略因子，並利用Fama-MacBeth兩步驟橫斷面迴歸法評估模型市場定價能力。結果發現定價能力均顯著，而利差交易策略之成因為股價指數波動因子(△EVOL)，因其可能連動匯率波動而呈現負相關；動能交易策略成因則為股價指數波動因子(△EVOL)及落後期限利差因子(△LTS)，主要因動能交易主要來自於市場資訊反應不完全，前者成因因子提供更大的動量執行交易策略、後者則因投資人在不同景氣循環下而有不同的投資反應，如景氣擴張的過度自信與樂觀、景氣衰退下產生行為財務領域中的處置效果，使兩成因與動能策略因子呈現正相關。 / This paper is based on the model of three-factors foreign exchange returns. So we test whether three-factors FX model which adds the factor of momentum can have stronger ability to explain currency excess return than two-factors FX model in the sampling period of February 1985 to October 2016. And the 37 kinds of currency are sorted by double sort method and become 9 portfolios. Finally, no matter coefficient of determination or residual error, three-factors FX model performs well. Further, we use stepwise LS regression (independent variable should have statistical significance in 90% confidence interval) to find which factor we choose can cause carry and momentum strategy profit in three-factors FX model. Next, using Fama-MacBeth two-step regression to estimate the asset pricing ability. The results represent that all contribution factors which get from stepwise LS method are significant. Carry trade strategy and △EVOL are negative correlation, because volatility of stock index will influence volatility of FX. And there have the positive correlation between momentum trade strategy and two factors(△EVOL and △LTS). Just because the profit from momentum strategy comes from the incomplete reaction of market information and △EVOL give more motive force. Besides, there have different investment reactions in diverse business cycle. Investors are over confident and optimistic during the period of recession and have disposition effect during the period of boom.

This collection of essays examines the topic of macroeconomic stabilisation in an international context, focusing on monetary policy, capital controls and exchange rates. Chapter 1, written in collaboration with Giancarlo Corsetti and Joao Duarte, reconsiders the effects of common monetary policy shocks across countries in the euro area, using a data-rich factor model and identifying shocks with high-frequency surprises around policy announcements. We show that the degree of heterogeneity in the response to shocks, while being low in financial variables and output, is significant in consumption, consumer prices and macro variables related to the labour and housing markets. Mirroring country-specific institutional and market differences, we find that home ownership rates are significantly correlated with the strength of the housing channel in monetary policy transmission. We document a high dispersion in the response to shocks of house prices and rents and show that, similar to responses in the US, these variables tend to move in different directions. In Chapter 2, I build a two-country, two-good model to examine the welfare effects of capital controls, finding that under certain circumstances, a shut-down in asset trade can be a Pareto improvement. Further, I examine the robustness of the result to parameter changes, explore a wider set of policy instruments and confront computational issues in this class of international macroeconomic models. I document that within an empirically relevant parameter span for the trade elasticity, the gains from capital controls might be significantly larger than suggested by previous contributions. Moreover, I establish that a refined form of capital controls in the shape of taxes and tariffs cannot improve upon the outcome under financial autarky. Finally, results show that the conjunction of pruning methods and endogenous discount factors can remove explosive behaviour from this class of models and restore equilibrating properties. In Chapter 3, I use a panel of 20 emerging market currencies to assess whether a model that combines fundamental and non-fundamental exchange rate forecasting approaches can successfully predict risk premia (i.e. currency excess returns) over the short horizon. In doing so, I aim to overcome three main shortcomings of earlier research: i) Sensitivity to the chosen sample period; ii) seemingly arbitrary selection of explanatory variables that differs from currency to currency; and iii) difficulty in interpreting forecasts beyond the numerical signal. Based on a theoretical model of currency risk premia, I use real exchange rate strength combined with indicators for carry, momentum and economic sentiment to homogeneously forecast risk premia across all 20 currencies in the sample at a monthly frequency. In doing so, the model remains largely agnostic about structural choices, keeping arbitrarily imposed restrictions to a minimum. Results from portfolio construction suggest that returns are significant and robust both across currencies as well as over time, with Sharpe Ratios in out-of-sample tests above 0.7.

本研究提出了一個類神經網路機制，可以及時有效的預測利差交易(carry trade)的收益。為了實現及時性，我們將通過Tensorflow和圖形處理單元(GPU)來實作這個機制。此外，類神經網路機制需要處理具有概念飄移和異常值的時間序列數據。而我們將透過設計的實驗來驗證這個機制的及時性與有效性。 在實驗過程中，我們發現在演算法設置不同的參數將影響類神經網路的性能。本研究將討論不同參數下所產生的不同結果。實驗結果表明，我們所提出的類神經網路機制可以預測出利差交易的收益的動向。希望這個研究將對機器學習和金融領域皆有所貢獻。 / This research derives an artificial neural networks (ANN) mechanism for timely and effectively predicting the return of carry trade. To achieve the timeliness, the ANN mechanism is implemented via the infrastructure of TensorFlow and graphic processing unit (GPU). Furthermore, the ANN mechanism needs to cope with the time series data that may have concept-drifting phenomenon and outliers. An experiment is also designed to verify the timeliness and effectiveness of the proposed mechanism. During the experiment, we find that different parameters we set in the algorithm will affect the performance of the neural network. And this research will discuss the different results in different parameters. Our experiment result represents that the proposed ANN mechanism can predict movement of the returns of carry trade well. Hope this research would contribute for both machine learning and finance field.

Cette thèse traite deux sujets distincts, les deux représentant des enjeux importants pour un grand nombre de Pays d’Europe Centrale et Orientale (PECO). La première partie porte sur les emprunts en devises étrangères. Plusieurs études antérieures montrent que dans de nombreux PECO, l’endettement en devises étrangères a augmenté de manière considérable avant la crise et est devenu un enjeu majeur pour les entreprises, les ménages et pour la politique budgétaire et monétaire. Pour évaluer les risques associés à l’endettement excessif en devises étrangères, nous étudions la volonté des entreprises d’apparier la composition en devises de leurs actifs et leurs passifs ainsi que leurs incitations à dévier de l’appariement parfait. Nos résultats fournissent des preuves solides à l’appui du rôle de la couverture naturelle. Néanmoins, ce dernier n’est pas le motif principal d’endettement en devise étrangères : le motif de couverture naturelle n’explique qu’environ 10 à 20 pour cent de la dette totale en devises étrangères des entreprises avant et pendant la crise, respectivement. La plus grande partie de la dette en devises étrangères correspondrait, au moins en Hongrie, à des positions de carry trade détenues par des sociétés non financières. La deuxième partie de la thèse est consacrée à l’exploration des liens entre les systèmes socio-fiscaux et l’offre de travail à la marge extensive. Le deuxième chapitre propose une nouvelle stratégie de modélisation de l’offre de travail comme alternative aux deux approches dominantes basées sur le calcul marginal et les modèles d’utilité aléatoire. Finalement, le dernier chapitre utilise ce modèle pour quantifier la part de la différence entre les taux d’activité tchèque et hongrois qui peut être expliquée par les divergences des systèmes d’imposition et de protection sociale. Les estimations donnent des élasticités d’offre de travail similaires, ce qui suggère que les préférences individuelles sont essentiellement identiques dans les deux pays. Nos résultats montrent que la moitié de l’écart entre les taux d’activité s’explique par les différences des systèmes socio-fiscaux. / This thesis deals with two distinct topics, both of them representing central issues for many Central and Eastern European (CEE) countries. The first part of the thesis focuses on foreign currency (FX) lending. Several previous studies point out that in many CEE countries, FX borrowing rose significantly before the crisis and has become a major challenge for firms, households and for fiscal and monetary policy. To evaluate the risks associated with excessive FX indebtedness, we investigate firms’ willingness to match the currency composition of their assets and liabilities and their incentives to deviate from perfect matching. Our results provide strong evidence to support the role of natural hedging, however, it is not the primary motivation for firms to choose foreign currency : it explains only about 10 percent of the overall corporate FX debt during the pre-crisis and 20 percent during the post-crisis periods. Most likely, the largest part of the corporate FX debt, at least in Hungary, corresponds to open carry trade positions held by non-financial corporations. The second part of the thesis is devoted to exploring the links between tax-benefit systems and labour supply at the extensive margin. The second chapter presents an alternative modelling strategy of labour supply to the two dominating approaches based on marginal calculus and on random utility models. Finally, the last chapter uses this model to quantify the difference between the Hungarian and the Czech participation rates that can be attributed to differences in taxation and welfare benefits. We find that the estimated labour supply elasticities for the Czech Republic are very close to the results for Hungary, suggesting that, at least in this dimension, individual preferences are similar in the two countries. Results suggests that about one-half of the total difference in the participation rates can be explained by differences in the tax-benefit systems.