Is gold's move parabolic?

[murphy_p_t]

I've seen more than one article claiming that gold has gone parabolic during its 12 year bull run. But I'm not so sure. Looking at the chart, say 5 year...it almost looks linear...http://www.kitco.com/charts/techcharts_gold.html

Or maybe some other geometric shape is dominant.

Anyone great at charting who can make a definitive statement on this?

[AgAuMoney]

I think you mean exponential (the traditional chart where the line goes from nearly horizontal on the left to nearly vertical on the right).  But maybe you do mean parabolic where it goes exponential, peaks and then falls down nearly as fast.  Either way the analysis is essentially the same.  The main difference is that exponential seeks to identify BEFORE the peak so that action can be taken before it is too late rather than after the peak when the pattern is clear but too late to do anything about it.

Generally to be exponential the growth has to be based on the previous growth.  In other words, some fixed relationship to the previous, like bacteria doubling, water lilies doubling, rabbits reproducing, etc.  That kind of exponential growth is what causes the crash -- the other side of your parabola -- resources needed to support the current size are exhausted, forget what is needed to support the rate of growth.

To properly detect something going exponential the vertical divisions need to be a non-linear scale (instead of 1,2,3,4...  it goes 1,2,4,8... or 1,10,100,1000).  This allows you to see the RATE of the change on the line instead of just the AMOUNT of the change.  (In other words, is the rate 10%, 9%, 8%, 7% or is the rate of change 10%, 10%, 10% or even increasing like 10%, 11%, 12%, ...)

In a way though, it all ties into the Mark Twain quote, "lies, damn lies, and statistics."  All of this stuff can be presented in a way to make any point the analyst cares to make.  There is a huge amount of judgment and decisions to make when analyzing and presenting any data series, so that even if you as the analyst are trying to be 100% objective, you have to filter the data and refine the presentation to get any meaning from it.  It is extremely hard to be 100% objective, so it is almost certain that subtle, subconscious bias is unavoidable.  And then if you are intent on making a certain point such that you intentionally manipulate the analysis and presentation...

[murphy_p_t]

AgAuMoney...that's what I was looking for...thanks.

Exponential. That leads to a nearly certain bust in short order. A mania.

Parabolic...this can go on for a long time? Bust is not certain nor imminent?

But as you pointed out, fitting the curves objectively is a huge difficulty.

[KevinW]

Exponential. That leads to a nearly certain bust in short order. A mania.

Not necessarily, any growth function based on compound interest is exponential. My money market accounts are growing exponentially.

[AgAuMoney]

Exponential. That leads to a nearly certain bust in short order. A mania.

Not necessarily, any growth function based on compound interest is exponential. My money market accounts are growing exponentially.

Exactly.  A slow rate of growth takes a long time to go bust.  It's kind of like the idea of making an investment at some reasonable guaranteed rate of return, but for an unreasonable time, e.g. 500 years.  If your ancestor would have loaned 100 pounds to the king of England 500 years ago for 5% simple per year, and assuming everyone until you let it compound and the current government honored that debt, they would owe you somewhere near 4 trillion (4x10^12) pounds (if I did the fv=pv*(1+r)^n correctly).

The first few years at 5% are no big deal.  But after 500 years of compounding, every year is a big deal.  If you were to start taking the interest instead of letting it compound, do you think you could live off of 200 billion (2x10^11) pounds per year?

See the "^n" in that fv formula?  That's the exponential.  fv is the future value, pv the present value, 'n' is the number of compounding periods where you receive the percentage rate 'r'.  E.g. 5% per year, r=.05 and 'n' is the number of years.  Not multiplied by the years, but raised to the power of the years.  Makes for fun investing fantasy, anyway. 

[Gosso]

If you look at gold on a log scale, as one should, then it looks fine.  Nowhere near as steep as the 70's.

[img width=700]http://i47.tinypic.com/2v99kbn.jpg[/img]
Source: FRED, www.tradingeconomics.com

Gold appears to be fairly priced (based on global M2, which I like to use, although not sure if it is appropriate).  Gold could easily hit $10,000 during a bubble, or fall to $500 if the Fed decided to pay +4% real interest on bonds and the economy was healthy/booming.

Plus as time moves forward, global M2 should also increase, which will help support the fair value of gold.  It seems quite plausible that gold could hit a fair value of $10,000 in 10-15 years.  But of course the market price could be much higher or lower.