|
Title: 2013-07-25 Konrad Graf: Tiger cub growing up? Bitcoin weekly closing prices Post by: n8rwJeTt8TrrLKPa55eU on July 25, 2013, 03:44:13 PM A more sober look at exchange rate volatility.
Quote Here is a chart of the all-time BTC/USD exchange rate on Mt. Gox presented differently than usual in several respects. First, it is built on weekly weighted-average exchange rates, which reflect not only moment-to-moment changes, but how much volume changed hands at which rate. Second, the dateline is not consecutive, but is for a single calendar year (2011 as calendar base). Each year has its own line with its respective weeks matched to the 2011 weeks within a couple of days. Third, a logarithmic scale is used. http://konradsgraf.com/blog1/2013/7/25/tiger-cub-growing-up-bitcoin-weekly-average-closing-prices-y.html Title: Re: 2013-07-25 Konrad Graf: Tiger cub growing up? Bitcoin weekly closing prices Post by: aigeezer on July 25, 2013, 03:55:53 PM Nice article. Useful data to help put the more common strident headlines into perspective. I liked the cautionary metaphor also:
"Still, all told, that little tiger does seem to be growing up. It grows larger each year. How cute and silly when a cub ambushes a rustling leaf and tumbles off the side of a fallen tree. Yet grown-up versions of such creatures hunt down and eat large animals. As the years go by, dismiss this still playful creature lightly at your own risk." Title: Re: 2013-07-25 Konrad Graf: Tiger cub growing up? Bitcoin weekly closing prices Post by: dree12 on July 25, 2013, 04:20:59 PM One does not simply "remove" exchange rate volatility by introducing weighted averages. Anything considered "volatile" can be tamed in this manner. Weighted averages over long periods are effective ways of determining the long-term trend. They do not measure "volatility", so this article cannot be presented as a "sober look at exchange rate volatility".
Investopedia: Quote Volatility, as expressed as a percentage coefficient within option-pricing formulas, arises from daily trading activities. Here's an extreme example to put this to rest. Look at the following chart. Is this "volatile"? Very much so. https://i.imgur.com/PDpBRKd.png However, once we smooth it out with averages, it seems all the volatility has disappeared: https://i.imgur.com/HXEHEnd.png Remember that the point of averages is to smooth out variance. Consequently, using averages to measure variance is absolute folly. Title: Re: 2013-07-25 Konrad Graf: Tiger cub growing up? Bitcoin weekly closing prices Post by: n8rwJeTt8TrrLKPa55eU on July 25, 2013, 04:39:27 PM One does not simply "remove" exchange rate volatility by introducing weighted averages. Anything considered "volatile" can be tamed in this manner. Weighted averages over long periods are effective ways of determining the long-term trend. They do not measure "volatility", so this article cannot be presented as a "sober look at exchange rate volatility". Ok ok, how about simply "a more sober look at exchange rates". Regardless of nitpicks in the terminology or methodology, I think the message of the article still stands: journalists and many in the Bitcoin community focus too much on short-term gyrations of the Bitcoin trees, as opposed to longer-term perspectives on the Bitcoin forest. Note, I'm not the author. But I do like his current and past work. Title: Re: 2013-07-25 Konrad Graf: Tiger cub growing up? Bitcoin weekly closing prices Post by: aigeezer on July 25, 2013, 05:36:48 PM The article provides a refreshing balance to the typical media piece that stresses a "dizzying climb" to the highest high they can cherry pick and an "abrupt plunge" to the lowest low they can find.
The gimmick of such media pieces is to depict time bounded by their hand-picked shock-value data points. Using year-over-year end points eliminates such trickery but does not itself speak to the issue of volatility within the year. The author reminds readers that the BTC price was at its extreme values for a matter of minutes or hours over the course of a year. Hugely important if you happened to be trading then, but effectively irrelevant otherwise. The typical media piece urges the reader to imagine that such wild price swings are always in play. Be afraid, be very afraid. Like the author, I find it useful to look at data using different levels of granularity. Each "lens" carries useful information and no single view carries the whole story. As always, you have to figure out what you're looking at and what its value is to you. Good article, in any case. :) |
