In a comment to an earlier post ( Market And Manipulation) I expressed the opinion that some of Hurst’s principles are applied differently with respect to forex pairs. The current analysis of the EURUSD is a good example. The generally accepted approach is to phase the waves at the lows and that the waves exhibit a simple harmonic relationship to one another by a factor of 2:1 except for the 54 month wave which is 3:1 to the 18 month wave. However, that approach is currently causing some difficulty in the EURUSD analysis.
Below is a weekly chart of the EURUSD with the 80 week and 10 week price waves shown at the bottom. The 80 week wave has been dominant for many years but since 2008 the filter output has been phasing to the highs (yellow arrows) as opposed to the lows.
Below is a weekly chart of the GBPUSD. I ran the exact same filter on the data to extract the 80 week wave. It shows the same very dominant 80 week wave but here the filter output is phasing to the lows (green arrows) and also very closely to the highs.
I believe this subtle difference in phasing behavior is due to the fact that the EURUSD has a perfect inverse relationship to the dollar index (which can be traded separately) whereas none of the other major forex pairs exhibit such a perfect correlation. The chart below is a weekly chart of the dollar index showing the 80 week wave based on the same filter. It shows the same dominance with phasing at the lows (green arrows) but not at all the highs.
Another important distinction with respect to forex involves the wave next longer than the 80 week wave. Hurst’s nominal model states that the next longer dominant wave is 54 months. Due to the fact that the 18 month wave is so dominant, one might be tempted to arbitrarily allocate every third low as a 54 month low. However a long term analysis does not provide for such a result.
Below is a monthly chart of the GBPUSD with the filter output of the 18 month wave over a 41 year period. The consistency of the wave is astonishing. Over a 41 year period it averages 18.2 months with very little modulation. The prominent,visually evident lows in the data do not appear to consistently separate into groups of three oscillations of the 18 month wave. The dominancy envelope is suggestive of a somewhat modulated 101 month wave subdivided into groups of five or six 18 month waves. I suggest care should be taken in applying Hurst’s 54 month wave of his nominal model to forex.
So where does that leave us? The good news is that the waves shorter than the 80 week wave (i.e. 40,20, & 10 week) exhibit a very close simple harmonic relationship to one another. The EURUSD is subject to some modulation but it is usually easy to pick up. Also the 10 week and the 20 week have synchronized at the major lows in the past. The 40 week will synchronize at the lows but not always. It will phase occasionally to the highs.. The chart below is a daily chart of the EURUSD with the filter output of the 10 week and 20 day price waves. It appears that the 10 week wave has turned down again and closed below the prior low close (1/23/15), which also was a 20 week price wave low (not shown). That suggests that the Euro is headed lower.
The Euro has been below its 10 week FLD for over forty weeks! Because of the difficulty discussed above in identifying the lows of the longer waves, forecasting when the low will occur is a bit problematic. History suggests that if the price goes lower, one of the next two 20 week lows has a high probability of being the low. However one must exercise extreme caution in trying to catch the proverbial falling knife or you may just catch it in your chest!