High Frequency Data and Volatility in Foreign Exchange Rates

摘要: Exchange rates, like many other financial time series, display substantial heteroscedasticity. This poses obstacles in detecting trends and changes. Understanding volatility becomes extremely important studying series. Unfortunately, estimating from low-frequency data, such as daily, weekly, or monthly observations, is very difficult. The recent availability of ultra-high-frequency tick-by-tick to large institutions creates a new possibility for the analysis volatile article uses foreign-exchange rates explore this type data. Unlike high-frequency data have high negative first-order autocorrelation their return. In article, I propose model that can explain estimator daily hourly estimates exchange rate show some interesting patterns.