This is pretty much how she drew the chart and said she thinks it is going lower before doing anything else.I have not said much about Silver or the Silver ETF (SLV). I have been clear that I think the market is tougher to assess and more volatile than gold. I do know the market is smaller for silver.
According to Portfolioscience.com the standard deviation for GLD is 22.89 and it is 40.20 for SLV.
In trying to find a good way to take on some commodity exposure SLV has more octane than I want to deal with.
7 comments:
Roger: They are different markets, for sure. Silver is also regarded as the "people's money" and has several industrial uses.
Central Banks, however, don't have big interest in it.
I have seen them trade in tandem, but the ratios do change and today teh ratio favors silver.
Silver is an extremely important industrial metal. It has the rare property of being more electrically conductive in its oxidized form. It is used a lot in electrical switching.
Steve Matthews had an article in the October 2003 (#23) issue of the Alchemist that I thought raised some interesting points: http://www.lbma.org.uk/publications/alchemist/alch_arch.htm
Russell
Oh, puh-LEEZE not a reference to the "standard deviation" out-of-context! We need two things for a standard deviation to be meaningful: a time frame and an average.
Assuming they calc'ed the standard deviations correctly, a STDEV of 23 or so, on a price of 58 or so, is about 40%. The STDEV of SLV being 40 or so, on a price of 110 or so, is about 36%. Therefore the volatility is about equal, with SLV actually being LOWER. That is, assuming the calc of STDEV is correct.
In isolation, it's meaningless. Over what period was the STDEV calculated? One year? One month? It would make a difference.
Looking at the current price in relation to the 14-day average true range, we get a ratio of 48 for GLD (relatively low variability) and about 26 for SLV (relatively high variability). So by this definition, SLV is more variable.
If one sizes the position based on variability, one removes the risk from the variability.
those were 3 month numbers
one other thing in building the screen (if that is the correct word) I did account for the dollar difference. I used 100 SLV and 200 GLD.
the standard deviation measured in dollars by porfolio science is $177 for GLD and $289 for SLV.
Is it calculated correctly? That is a good question. I have no reason to think otherwise, but....
Further as I look at a chart that compares the two, SLV looks much more volatile to me. That is hunch and my faith in portfolio science backs up the hunch. FWIW.
Well, you gotta understand us math nerds are gonna ask questions so we need some stuff spelled out. If you were posting a STDEV adjusted for the dollar difference, you should have said so. It would have made more sense to me, and probably some other readers went "huh?" at it. I'm still going "huh?" at a $177 standard deviation on a $60 issue. Are you missing a decimal? What's the link where you got that?
Three-month numbers? I think that's too long to be responsive. Lot can change in three months.
The closing price to ATR ratio is a good measure, in many ways better than the STDEV in my opinion, and it places SLV at more variance than GLD. To normalize the risk, I would size a speculation in SLV at just over half the size of a GLD position.
math nerd, hehe
the $177 is based on a position of 200 GLD as I mentioned was the inputas the $289 was for 100 SLV.
There is a one month calculation on portfolioscience that is not, IMO, dramatically different.
Post a Comment