USD Index Long Term Analysis - pinned topic

Posted this analysis earlier in the year.

Shared these charts privately in August.

So a pullback in gold/silver is no surprise as we finish 2016, and based on this look, a bearish period to come for the $USD. More short term analysis to follow.


US Dollar at an 18 month peak?

Eurusd i suspect it still on moving down to form f fld until 0.8. Thus very similar pattern found in dollar index dxy. I do not think eurusd can go longer than 20 years before it completely wipe out. I use 1 day timeframe to do analysis. My data is from year 1975.


Here is a quick and dirty look at a filter analysis of that DX data. It clearly shows a dominant 4 year wave over decades with a recent duration fluctuation. The envelope appears to tracking a 16 year price wave whose down trendline has been broken. The chart is identical to the chart of the Euro, just inverted.


1 Like

The USD Index is on the cusp of doing something really important in the coming weeks/months. I will probably be pilloried for posting this analysis and these thoughts. There is no denying a ~16 year cycle for the USD as shown by William’s bandpass filter, many currencies and not surprisingly gold. So when I look at the USD index, or the CDN dollar, I treat it like gold. Thus I use my favored nominal model for gold when analyzing the USD index. And that is not Hurst’s nominal model. The gold model involves a near 16 year cycle, 7.5 year, 45 month and so on.

Here is a long term view showing the 4 year cycles. A comparison to William Randall’s chart shows pretty much the same cycle. I’ve run Sentient from the start of the data set in 1967 using a gold nominal model. No pinning at all here. You be the judge.

So now we move to the larger cycles and we focus on the 7.5 year, ~ 15-16 yr, and then the 2:1 harmonic up from that which is ~ 30 years. Remember how we suggested that gold & silver probably had a ~30 year cycle? Same idea applies here IMO.

So now I am going to show the controversial chart. What is the nature of the high being formed, that is, if a high is being formed for the $USD? Cyclewise, it would be a 45 month, 7.5 year, 15 year, and 30 year synchronous peak. That’s right. It does make some sense given the weak rise we’ve seen since the 2008 low. The USD index did break the 30 year trendline to the upside, which I will call the 30 year VTL. And I have been waiting for this great surge in the USD since and yet we are still waiting. Some would say the USD index is in a pause zone or forming a continuation pattern. It is always possible and I don’t discount it. But look at what Sentient is saying. It has 50 years of data and this is what it comes up with? Why??


So am I convinced that the 15-16 year cycle is in for a top in the USD Index? That is a question mark and long term cycles can have a wide range. This cycle has had 15.4 and 16.4 year intervals from the 1970 - 1985 and 1985 - 2001 peaks (average 15.9 yr). The current cycle is 15.2 years along from 2001 to present. So this cycle is a bit short. Sentient marked the last cycle peak at 14.4 years. Again I am a little skeptical that the high has been seen but it is possible. For the low to low cycle intervals, we have 13.9 and 15.6 years for the 1978 - 1992 and 1992 - 2008 lows (average of 14.75 yr).

The thing to be aware of is that these cycles are very during. Once a trend reverses, then these 15-16 year cycles typically turn years later at projected lows or highs.

Here is the Hurst Cyclic Model Summary

Food for thoughts

So to conclude this analysis, here is a possible technical explanation of why we might be looking at a major top in the $USD Index. I don’t know how this very long term chart was constructed. What I have studied are the major peaks (15 and 30 year for example). The thing to notice is that there is a trend line, which is drawn with a 150+ year duration. It is curious that the $USD Index has been playing with this trendline for a number of years. Is it real? Well if Sentient is correct and a major top is forming, then this crude USD chart gives some technical understanding of why it might be happening. So we have to look to other things like individual currencies, gold and silver and commodities to assess if this could be happening.

So I’ve put it out there. I don’t know if this view is correct, but personally we have been selling $USD back into $CDN this year in our accounts at every good opportunity.


Sorry here is the long term chart and line.

BTW, I thought of blog posting this at but I will wait until I see more evidence that this is a possible scenario.

Hi John.

That’s a nice analysis. It is nice to see that ST’s algorithm and the my filter approach show approximately the same 4 year wave. My 16 year wave approximation is also the same as ST’s. The chart below is a technique I like to use for projecting the waves into the future. It is a combination of a “composite” envelope and a slight variation of Hurst’s non-realtime envelope technique from Profit Magic. The price levels are inaccurate but the timing of when the envelopes become congruent, which indicates a turning point, is very accurate.



John and William,

I have been working on an analysis of the dollar and stock markets also and figured I’d chime in.

Below is a rough phasing analysis of the DJIA since 1945 versus a price chart of the Fed’s trade weighted dollar (not DXY but similar). Spectral analysis was used on the DJIA and the results are consistent with the long-term phasing in Chris Grafton’s book and David’s 9/16/16 webinar using Hurst’s original nominal model. I have also investigated shorter nominal models that fit the last 25 years better but the original model seems to still work well over the long term. If you use the second low after the stock crashes in 1974 and 2009 (both 9 year lows), it helps resolve some issues.

-The last two major peaks in the dollar have been associated with 18 year troughs in the DJIA.
-The 1/2016 peak in the dollar and 3rd 54 month trough in the most recent DJIA 18 year cycle line up with the 8/1998 peak in the dollar and 3rd 54 month trough in the previous DJIA 18 year cycle.
-The dollar may have significantly more upside.
-The high in the dollar and 18 year low in the DJIA should be in early to mid 2020.
-This is not inconsistent with William’s channel analysis showing the second high for the dollar in this same time frame.

Because of the enormous amount of borrowing in dollars by foreign governments and corporations globally. a strong dollar significantly tightens global financial conditions, which in the past has led to trouble.




I have been a USD bull since the 2011 top in gold/silver. However, time seems to be running out for the DXY. The dollar index is up 8.5 years now from the 2008 low. Granted this is the longest dollar rally versus the 80s (~6.5 yrs) and 90s (~8yrs). There are some very large cycles at play here so I will not be betting against them. Namely the 8 and 16 year cycles.

The question is whether this pattern which started from early 2015 is a consolidation (pause zone - Hurst), or a topping pattern (eg. triple top). I don’t have any evidence as to which way it goes other than cycles say time is running out. The other thing I would suggest is that it would be odd to have a pause zone form here, precisely where an 8 and 16 year cycle high should arrive for the USD. I could go on from a cycles point of view, but I think I have expressed my views fairly clearly earlier in this thread.

From a purely technical point of view, the DXY index tested the 100 level in early 2015 on high volume. This is the third test now on lower volume with weakening internals. The month is not over so TWT.


PS The cycle bubbles drawn are 7.66 years each top to top, and the next cycle up is 15.3 years. We are in the time band for the 16 year cycle to turn (ie. top).

1 Like

A near term pull back is very likely, IMO. Resistance/volume confirm this. But according to the peak analysis on the longer data set from the 1973 start date, the weakness in DXY could only be temporary (similar to the 1999 pause). Could be a late 2017 or early 2018 peak (16.5 year cycle-12/1984, 6/2001, 12/2017). Keep in mind, there was a second peak for DXY in 2002, so the dollar could stay strong for another year after this projected high.

It makes sense. The Fed is tightening and everyone else is still doing QE. This rate/growth differential will be aided by the myth of Trumpenomics (at least near term). There is $10 trillion outstanding in the dollar carry trade, most of it in EM (source; BIS). A dollar rally of this magnitude will cripple foreign borrowers in dollars (casualty #1 was Brazil). I compare the period from the June 2014 low in DXY culminating in the Jan/Feb stock selloff to the 1997-1998 Asian financial crisis. Just an appetizer for what is to come. The pain trade. Lines up very nicely with Hurst cycles and the 18 year low for stocks projected in either 2019 or 2020.

I have extracted pertinent analysis on the USD Index and pinned the thread. I have closed the thread to new posts and will leave it as reference for the coming year.

Just wanted to updated this long term analysis. When I posted in Nov. 2016, I was not convinced the peak was in even though Sentient was looking for it. I’m still not 100% certain the high is in for the $USD. But what hasn’t changed is that the same message is still there for a major top …