This research investigates the ability of genetic programming to build profitable trad- ing strategies for the Foreign Exchange Market (FX) of one major currency pair (EURUSD) using one hour prices from July 1, 2009 to November 30, 2012. We rec- ognize that such environments are likely to be non-stationary and we do not expect that a single training partition, used to train a trading agent, represents all likely future behaviours. The proposed adaptive retraining algorithm – hereafter FXGP – detects poor trading behaviours and trains a new trading agent. This represents a significant departure from current practice which assumes some form of continuous evolution. Extensive benchmarking is performed against the widely used EURUSD currency pair. The non-stationary nature of the task is shown to result in a prefer- ence for exploration over exploitation. Moreover, adopting a behavioural approach to detecting retraining events is more effective than assuming incremental adaptation on a continuous basis. From the application perspective, we demonstrate that use of a validation partition and Stop-Loss (S/L) orders significantly improves the perfor- mance of a trading agent. In addition the task of co-evolving of technical indicators (TI) and the decision trees (DT) for deploying trading agent is explicitly addressed. The results of 27 experiments of 100 simulations each demonstrate that FXGP sig- nificantly outperforms existing approaches and generates profitable solutions with a high probability.

This dissertation comprises two essays. The first essay develops and tests a theoretical model that provides new insights when markets are partially segmented and the purchasing power parity (PPP) is violated which seems to be the case for the majority of national markets. The theoretical part derives closed form solutions for asset prices and portfolio holdings. Particularly, we show that deviations from PPP in mildly segmented markets induce a new form of systematic risk, termed segflation risk, and in equilibrium investors require compensation for this risk. A strong feature of the model is that it provides a theoretical framework for testing important issues; such as, pricing of foreign exchange risk and world market structure. The model also nests several existing international asset pricing models and thus provides a framework to distinguish empirically between competing models. The empirical part of the essay provides an empirical validation of the model for eight major emerging markets. The results give support to the model and point to the importance of the segflation risk which is statistically and economically significant. / The second essay uses our theoretical model to address the question of whether the IFC investable indices are priced globally or locally. Indeed S&P/IFC provides two emerging market indices: the IFC global index (IFCG) and its subset the IFC investable index (IFCI). Since the IFCI is fully investable, both the academic and practitioners implicitly assume that this subset of emerging markets is priced in the global context. This is a critical assumption for corporate finance decisions and portfolio management. Hence, this essay investigates the pricing behavior of the IFCI index returns using a conditional version of our model that allows for segmentation and PPP deviations. The results suggest that local factors are important in explaining returns of the IFC investable indices and that the return behavior of IFCI indices is similar to that of the IFCG.

This thesis presents three essays on the role of international capital flows in growth, real exchange rate behavior and the conduct of domestic monetary policy in four Asian economies. The first chapter develops an endogenous growth model based on an infinitely-lived optimizing representative agent. Data from the four Asian countries is used to test the implications of the model. Using applied time series econometric techniques, the results for Malaysia, Philippines and Thailand lend credence to the endogenous growth process, while it is rejected for Indonesia. Chapter 2 develops a three-good model for the internal real exchange rate to identify the fundamental determinants of the internal real exchange rates for exports and imports. The examination of the time series properties of the variables suggests that the internal real exchange rates in the ASEAN-4 countries were indeed driven by the fundamentals derived from the model. Furthermore, the results indicate that there was no misalignment between the actual and equilibrium real exchange rates. Movements in the real exchange rates were thus equilibrium responses to changes in the fundamentals. The third chapter estimates coefficients of capital flow offset to domestic monetary policy and sterilization and analyzes the implications for domestic monetary autonomy. The relative performance of the monetary model and the portfolio balance model is compared using quarterly data for the four countries. The empirical results show that the capital flow offset was less than complete and that sterilization turned out to be ineffective in three of the four countries.

Since the early 1970s, exchange rate fluctuations have characterised the behaviour of the external value of many currencies in both high- and low-income countries. Up-and-down movements in real exchange rates have been observed under fixed as we:ll as flexible arrangements. This is in spite of the fact that many less developing countries (until the 1980s), unlike the major industrialised countries, opted to retain relatively rigid exchange rate systems after the collapse of the Bretton Woods system. Exchange rate volatility has been a subject of much concern in government, business and academic circles because it has been associated with negative effects on the performance of developing economies. Consequences of these large swings in exchange rates have included uncertainty and delays in business decisions, resource misallocation, interest rate volatility and real exchange rate misalignments. For the period, froln1970 to 1996, this study investigates the phenomenon of real exchange rate disequilibrium in developing countries, with an empirical and econometric examination of South African data. Using the ordinary least squares and the EngleGranger cointegration techniques, this investigation found that government consumption of nontradables, the price of gold in rand, the overall terms of trade and the rate of depreciation are important determinants of the short-run behaviour ofthe real effective exchange rate in South Africa. With regard to the long-run the permanent componen1ts ofthe fundamentals - namely, technological or productivity improvement, trade policy, governm1ent consumption of nontradables, disposable income, capital flows, the terms of trade excluding gold and the rand price of gold -, were found to be significantly related to the equilibrium conduct of the real effective exchange rate. Instances of real exchange rate misalignment were found in both periods of fixed and flexible exchange rate management. / Thesis (M.Soc.Sci.)-University of Natal, Pietermaritzburg, 1999.

Currency exposure has become a widespread issue as more corporations of all sizes source and sell in overseas markets and compete both at home and abroad with international companies. Very few companies are unaffected by currency risk, whether directly or indirectly. Businesses that source products from foreign countries face the risk that exchange rate movement will erode gross margins if competition prevents selling prices from rising in tandem, while resource-based companies face the uncertainty associated with the fact that the world's commodities markets are denominated in US Dollars or Pounds Sterling while their costs are often denominated in their local currencies. Businesses that ignore exchange rate volatility expose themselves to unnecessary risk, which could have significant consequences if exchange rates suddenly move unfavourably. The volatility of the South African Rand over the past few years is forcing treasurers and other managers responsible for international trade to look anew at how South African exchange rate fluctuations affect their company's results. Many companies have suffered from the effects of fluctuating exchange rates; some have reported losses running into millions of Rand. While more and more firms realize that they should manage foreign exchange risk, not all of them have come up with an appropriate management strategy. There has always been a great deal of debate over the best approach to hedging, or the best methods to forecast exchange rates; however hedging is of the utmost importance for companies. With the recent volatility of the rand, the multinational firm covered in this thesis, showed foreign exchange losses amounting to several millions, using forward exchange contracts to cover its high foreign exchange exposures. The major disadvantage of the forward contract as experienced by the firm and shown in this thesis is that it is a legally binding agreement and thus the firm was bound to accept the agreed exchange rate and also the fact that the exchange itself had to be done. If the commercial reason for the exchange disappeared, the cost of cancelling the forward contract would be quite high. In addition, if the exchange rate at maturity was more favourable to the firm than the one agreed to in the forward contract, the firm will still have to honour the contract and will not be able to take advantage of the favourable exchange rate. Thus, with FEC there is the elimination of the opportunity for profit, should exchange rates turn out favourably. When purchasing a currency option, however, the holder is protected from downward movements in the exchange rate whist still having the opportunity to benefit if the currency moves favourably. Hence, the purpose of this thesis was to evaluate the use of currency options as an alternative hedging strategy to forward exchange contracts to manage the firm's foreign exchange risk. It was found that, had the firm used currency options as compared to FEC over the last four years, the firm would have made significant saving in spite of the option premium. The firm would have enjoyed the flexibility offered by currency options, that is, to let the contract lapse when it would not be to the firm's advantage thus making a lower payment for its imports than would be paid under the forward exchange contract for the same period. The results were tested over a period of four years to prove that the difference in payments using the FEC and the currency options were statistically significant. What was apparent from the research, however, was that though the multinational firm could choose from a vast array of financial instruments and currency derivatives to manage its foreign exchange risk, the firm chose to stick to using forward exchange contracts. The reasons varied from fear of dealing with the complexities of the many instruments available on the market to the limited resources within the foreign exchange department to understand the technicalities of the various instruments. The investigation revealed though forward cover as used by the firm was more efficient in terms of ease of use. Currency options when applied to cover the firm's foreign imports resulted in less cash outflow, making it better and more profitable than forward exchange contracts. Options contract, though more expensive, would have allowed the firms to let the option lapse and therefore benefit from spot exchange rates if these were more favourable. / Thesis (M.B.A.)-University of KwaZulu-Natal, Pietermaritzburg, 2006.

The immediate objectives of exchange-rate policy should be stability of output, stability of the rate of inflation and stability of the exchange rate itself. Moreover, exchange-rate policy is likely to be of some use in achieving these aims, as the modifications to the rational expectations theory which make monetary policy effective do so for exchange-rate policy as well, and additional channels of effectiveness operate in an open economy. There are various explanations for the volatility of exchange rates under free floating, but a more realistic picture of the reaction to external disturbances is obtained if step changes in interest rates are allowed for by considering the term structure of interest rates and the influence of the terms of trade on the demand for money. Capital is not in fact perfectly mobile internationally, which widens the range of policy options, and some modifications of the theory are required to study the determinants of capital flows in an imperfect world. One can use the correlation between deviations of output and inflation from trend to study the origin of disturbances. This is illustrated by a study of eighteen sub-Saharan African countries. Furthermore, the type and origin of disturbances have implications for whether some sort of dual exchange-rate or dual interest-rate system is desirable. Finally, a model of an open economy is presented in which exchange-rate policy can be analyzed whether specified in terms of exchange-rate targets or the degree of intervention, and with the possibility of a restricted forward foreign-exchange market. A variant of the model is estimated for South Africa for the period 1974-1981 and various exchange-rate policies are simulated.