CAD and oil hour-scale correlation: it's safer to rely on CAD |
| Written by Mikhail Kopytine | |
| Wednesday, 14 October 2009 17:42 | |
|
In the recent forex/CFD data, USD/CAD is negatively correlated with light oil (WTI) CFD. This is the same as saying that CAD, one of the commodity currencies, is positively correlated with oil. This is old news. In this article I take a deeper look at the issue and analyze the shape of the correlation peak. Analyzed on the hour time scale, the correlation peak is broad and somewhat asymmetric, indicating that it is much safer to rely on the guidance of USD/CAD in predicting the oil price, rather than other way round. The necessary caveat is that this is a time-integrated picture, covering a period from late August 2008. Fig.1: Historical bar charts of the data used in the analysis. Top: USD/CAD, bottom: light oil CFD. Time scale is hour. Time axis is labeled in MM-YY format. Fig.2: comparing volatilities of hour-by-hour logarithmic returns in USD/CAD and light oil CFD. The comparison is fair since the absolute levels of price are taken out of consideration in the logarithmic returns. The logarithmic return distributions on the hour scale have the usual quasi-triangular form, indicating power-law style distributions. Oil is a lot more volatile than CAD.
Fig.3: Cross-correlation of USD/CAD and light oil CFD, derived from the hour-by-hour logarithmic returns. Fig.3 presents the cross-correlation of USD/CAD and light oil over the time lag (hours), for round-the-clock trading hours and for 9am to 9pm New York time. A comparison with the same analysis performed repeatedly on the random data designed to mimic volatilities of USD/CAD and light oil lets one estimate the accuracy of the correlation measurements. The statistical reference, uncorrelated noise, is obtained for each particular time lag bin by analyzing 20 independent simulated pairs of uncorrelated time series, having average volatilities of USD/CAD and light oil CFD. The reference is presented as mean plus-minus 1 RMS, where RMS characterizes the distribution of the correlation value. A feature of Fig.3 (bottom panel) is non-flat (although quite predictable) behavior of the noise level with time lag, caused by the constraint on the time lags associated with the definition of the trading session time window. This can not be ignored otherwise one risks over-interpreting the picture. The area around zero is fairly safe since the noise is at the minimum when the lag is at an integer number of days. The noise is flat in the top panel where round-the-clock data are used. Naturally, as the random model responsible for the noise (red background in the figure) does not contain any correlation between the two exchange rates, it shows no correlation feature at the zero time lag. The overall peak asymmetry is to the left, into the area of negative lags (whereby light oil CFD would be lagging behind USD/CAD). The reason negative lags are interpreted this way is the definition: the lag td is defined as td = t1 - t2, where index "1" denotes USD/CAD and index "2" denotes oil. Therefore, negative correlation value at negative lags means that movements of the opposite direction in USD/CAD and oil happen earlier in USD/CAD, or CAD is leading and oil is following. Even though the tail of the correlation is directed to the left, the 1 hour time lag also shows a strong correlation indicating that changes in oil price can also lead USD/CAD. But it is a lot safer to play USD/CAD as an indicator for oil, rather than oil as an indicator for USD/CAD, since already the +2 hour bin is seen to contain a correlation of the opposite sign. The data used are from the period 2008-08-21 to 2009-10-14. |
Danica | next day forecast | 9am Eastern time | 14 forex pairs
Very nice and timely work here. Thank you very much for looking at this for me. Now we will get 4 or more times more trades with the cad related pairs.
I would also ask you? Why would usd/cad have anything in the world to do with Crude oil. My buddies suggest it is the conspiracy theory in action, ha ha!