Sözlü müdahalelerin para birimlerinin birlikte değişimlerine etkisinin ileri bilgisayar ortamında test edilmesi

In the last decade, both Federal Reserve System (FED) and European CentralBank (ECB) abandoned direct market interventions and relied on communication as their main policy toolto affect exchange rates. This paper investigates the impacts ofofficials' statements (oral intervention) on the covariance of theEUR/USD and JPY/USD. Using generalized autoregressive conditionalheteroscedasticity (GARCH) model's diagonal vector error correction(DVEC) representation, we find that strengthening oral interventionsin US and Japan decrease while in Eurozone increase the covariancebetween EUR/USD and JPY/USD. Also reversely, weakening oralinterventions in US and Japan increase while in Eurozone decreasethe covariance. Since oral interventions are explanatory variablesof the conditional covariance structure of G3 currencies (USD, EURand JPY), ignoring oral interventions may cause errors in foreignexchange (forex) covariance forecasts. During the estimationprocedure, we use a different approach than the commonly practicedin the literature. We solve the resulting optimization problem frommaximum likelihood estimation (MLE) of DVEC model in two steps:first by genetic algorithm (GA) and then by sequential quadraticprogramming (SQP) algorithm. Furthermore, to land at a better localoptimal, the experiments are conducted in NEOS Servers. Comparing our results withthose of benchmark S+ GARCH module (a commercial software), we findthat our approach yields much higher objective value than thebenchmark does. Hence, we conclude that our computationalmethodology provides substantial improvement to in-sample forexcovariance forecasting.Our results have applications in portfolio management as well