A Shocks & Surprises Approach to Currencies

Currency Conclusions

Strengthing Trends

• US Dollar: Since early January, the dollar has been on a strengthening trend vs the SDR, and that trend is not currently under threat. 
• Euro: The current strengthening trend has been in place since mid-September, and is not under threat.
• Rmb: The current strengthening trend has been in place since early October 2012, and is not under threat. 

Weakening Trends

• Yen: The weakening trend has been in place since mid-December and is not under threat. 
• Sterling: A new weakening trend emerged in mid-January, and is strongly in place. 
• Gold: Reflecting the dollar's strength, since early January gold has been on weakening trend, and that trend is not under threat.

On Watch

• Australian Dollar: The strengthening trend which was in place since the second week of January is now under threat. 

Argument & Methodology
Forecasting exchange range is for mugs. Despite it being a lucratively-rewarded profession, no-one knows how to forecast currencies with near-certainty or accuracy, and for good reasons. The first good reason is that the spot price will always reflect a fluctuating arbitrage between short-term cashflows and longer-term balance sheet necessities, and both sides of that arbitrage are fundamentally in the process  of new discovery, as is the appropriate arbitrage between them. For example, that current cashflows may support the Yen or the Euro can be a situation which can quite easily co-exist - for decades - with a situation in which balance sheet considerations look distinctly unappetising.

The second reason is simpler: for major currencies, fx markets are probably the nearest thing we have to perfect competition,  the logical extension of which is that your explanation is likely to be just as good, or bad, as mine, no matter how hard you (or I) work at it.  Competition, after all, exists to discover information which can be discovered in no other way.

But the itch, the desire to say something useful about currencies, remains. Even working day since Friday December 1st 1995, I have collected OANDAs' tick-averaged fx for all major Asian and European currencies, and consequently have a reasonable database from which to make observations. In practice I use them not to forecast, but rather construct from them rules-of-thumb about the state of individual currencies: are they stable, strengthening or weakening, and at what point to they look like breaking trend?    The rule for this is simple: I track moving averages and standard deviations measured from the point of inflection in 90-day trends. (Each day, the average and standard deviation will change as the trend is extended - typically as the trend lengthens, the average deviation from it will fall, whilst the standard deviation grows). Historically I find that when daily prices are more than two standard deviations away from that trend, a trend inflection usually follows.  

What I am attempting to do is simple and limited: I am trying to accurately identify current trends, and identify plausible threats to them.   I intend to publish my conclusions every Monday.