Proceedings of the International Conference on Advanced Research in Business, Management and Economics
The relationship between a Unified Financial Condition Index and the most actively traded USD based Foreign Currency pairs
Various regulatory bodies in the US use proprietary financial conditions indices as tools to measure the health of financial markets. While they share some common variables, they also differ in areas such as data frequencies of their respective models. The aim of this study is to propose a unified financial condition index centered around the most popular financial conditions indices used in the US and tests its relationship with the most actively traded USD based foreign currency pairs. Using weekly data over 1993-2018, this paper proposes a unified financial condition index (UFCI) under a principal component analysis framework. The index captures 78% of the variability inherent in St Louis Federal Reserve Financial Stress Index, the Chicago Fed National Financial Condition Index and the Adjusted National Financial Condition Index. Significant p-value of UFCI, homoscedasticity and a relatively stable root mean squared errors was observed only for EUR/USD. Mixed findings found as lags were increased suggests a weak relationship between UFCI and foreign currencies. The UFCI forecasting model is compared with the VIX (volatility index) based model, and also a random walk model. Although the UFCI model was superior only for the Canadian dollar, Chinese yuan and Indian Rupee after considering heteroscedasticity in errors, results were sensitive to number of lags and insignificant p-values.
Keywords: financial conditions, exchange rates, forecasts, principal component analysis.