Earlier today, Gary Evans directed me to this chart.
By coinicidence, CNBC’s Ron Insana featured a very similar chart this afternoon:
click for larger graph
The chart suggests that Gold has a lot higher to go . . .
Source:
Gold/Oil Ratio Extremes 3
Adam Hamilton
Zeal, August 19, 2005
http://www.zealllc.com/2005/gorex3.htm
The correlation between these two series is only 0.5.
I know, the chart looks a lot stronger.
Over the last 15 years the correlation between the price of oil this month and the price of copper has been 0.97 while the price of oil this month has a 0.93 correlation with the prior months price of copper.
Whats even scarier is that this chart also suggests oil has a lot further to go before it hits its peak.
If this price hurts, what about a 1980’s level price?
Correlations on time series like this are not valid, nor are stochastically-implied measures such as standard dev etc. This is exactly the kind of analysis that got LTCM into so much trouble.
Correlations on time series like this are not valid, nor are stochastically-implied measures such as standard dev etc. This is exactly the style of analysis that got LTCM into so much trouble.
D. Wallener,
I think that you should explain just why correlations like this are not valid.
I agree that a correlation never proves a causal connection between two variables, but it is suggestive of a causal connection, and it is behoving on you to show why this is not the case. The correlation is there, Why won’t it continue?
Having posed the question let me offer one answer which is nothing to do with oil or gold. It is that markets must be random otherwise half of the participants will become trillionaires and the other half will become destitute in a very short time.
It is usually the case that once everybody is aware that A+B=C, and once everybody places their bets accordingly, then A+B no longer equals C. The way that this paradox is often resolved is that A+B=C proves to be true in the end, i.e. the experts were correct after all, but the financial rewards only accrue to those who get the timing right.
It’s a very bad time for me to suggest that gold is overdue for an upward move because any idiot can make this claim after the last weeks gold appreciation. However, as someone who has invested in gold for the last four years, and in the absence of any evidence as to why the gold/oil disconnect is irrelevant, then I wonder what is more likely. Is the magic refinery fairy going to increase refining capacity to process the additional crude oil provided by the crude oil fairy? Or is the massive and growing US deficit fuelling a flight to the only true store of value, gold.
sheesh. you guys are really smart.
I thought he posted that chart because it’s pretty.
jw
P.S. real OR imagined, the relationship is due to inflation.
how does platinum figure into this picture?
how has the relative traded amt of platinum vs. gold changed over time?
Real Gold versus Real Oil
Don’t let nominal price comparisons fool you. Gold is still (relatively) cheap….
LTCM!!?? It was my understanding that LTCM was using a nobel-prize-winning mathmatical formula that theoretically indicated an investment’s “risk” and that by finding two investments with opposite risks, one could minimize the risk of each investment. That hardly compares to looking at charts and drawing crude conclusions, but what the heck do I know? Seriously, that’s was what I thought they were doing at LTCM, but probably everyone posting to this site has more Economics training than my little B.A. Please correct me if I am wrong.
Interesting chart – very useful for placing things in perspective and that is all it can do. Sorry guys, but you cannot forecast prices!
And why correlation analysis is valid or not? All it does is use historical data and ‘implied’ (forecasted) data to help identify a convergence/divergence trend. Just a more complex way of attempting to predict the future – the nearer the future, the more accurate.
My favourite comment is the inflation one -> the same factors that affect oil will most likey affect Gold also – but then there are different markets for each and varying factors. Oil’s future supply is questionable whereas most of the world’s Gold ever mined is still aboveground (i.e. it is recylcable where oil is perishable). Oil’s main market is transportation fuel whereas Gold’s is Jewelry (check gold.org).
… and the list goes on…and on… much like a Mandelbrot set really (and they make prettier charts).
So, if you want to work out whether gold has a lot further to go, get all the information you can (more than a chart is a good idea), do the opposite of what your broker tells you and avoid share market games.