
doi: 10.2139/ssrn.998584
This paper first examines the properties of the realized volatilities of US Dollar/Canadian Dollar spot exchange rate covering a time span of about three years and then the deseasonalized volatilities are estimated and forecasted using a fractionally-integrated model. The key feature of the realized volatilities is that they are model-free and also approximately measurement-error-free. More strikingly, the distribution of logarithmic realized volatilizes is approximately Gaussian. Our observations in this paper are consistent with previous studies. Usually a U-shaped pattern of the intraday volatilities should be observed due to opening-closure effects in the global market. We do not see a typical U-shaped pattern in the intraday volatilities for US Dollar/Canadian Dollar. The reasons are given in this paper. Because the intraday volatilities generated using the high frequency data shows this kind of intraday seasonality, we do a seasonal adjustment before estimation. After seasonal adjustment, we use ARFIMAX model to estimate and forecast the deseasonalized volatilities and the results are promising. And the future research will focus mainly on designing a ready-to-use automatic trading model based on the predicted volatilities in this paper.
| selected citations These citations are derived from selected sources. This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | 0 | |
| popularity This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network. | Average | |
| influence This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically). | Average | |
| impulse This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network. | Average |
