May 22 update: cited in Morningstar, WSJ network's MarketWatch, RealClear, Yahoo! Finance, and many other fine news outlets. Incidentally the previous article received from all sources, >200 facebook and >100 twitter shares, including the top picture on the viral #HowOldRobot. Last, as of May 22 the streak is now at 8 days instead of 7 (less than a once-annual event as has been the case for more than the past couple of years). Again, looking for S&P below 2103.
A 1% decline is -to some degree- a rather arbitrary level. But it's a small enough level that we typically see the markets fall on a later date by at least this amount, from any given or current day's intraday high. For this "chronically up" market, May 11 was the last time this occurred (intraday high of 2118 followed by a next-day low of 2085). From May 12 onwards -through today (May 21)- we have gone 7 days. During which time the markets on a later date have still not fallen at least 1% from (any of) those 7 intraday highs.
A streak of 7 trading days is also a lengthy period of time. During the past 2 years (>500 trading days), this streak of >7 days without an eventual decline of at least 1% has only happened 3 times. Otherwise there have only been a handful of times when the streak was capped at just 3 or fewer days, again without this 1% decline. This all implies an exceptionally, low probability event.
The recent market data, which clearly may not repeat in the same way going forward (only dim-wits construct alarming predictions: here, here, here), simply indicates to us that our current low volatility streak is rare. Generally occurring only once annually, and when it does it typically only lasts about 9 days: so about another day or two to go!
Much more likely though (say a >2/3 probability or so), we are just not in a record streak, and that on a future date we would have at least enough of a correction to contract this current market pattern back to its historic norms mentioned earlier. If this were true, then the May 15 intraday high of 2124 must see an eventual price of at least 1% less, so that the historic streak is capped at the otherwise typical 3 days noted above. Also, a 1% decline from this S&P of 2124 equals 2103 (a level that is a 1.3% discount from today's close of 2131). Stay tuned!
Incidentally, for those interested in an auxiliary topic of market-drop convolution, please refer to the treatment given in this article, or search for other related topics on the blog.
A 1% decline is -to some degree- a rather arbitrary level. But it's a small enough level that we typically see the markets fall on a later date by at least this amount, from any given or current day's intraday high. For this "chronically up" market, May 11 was the last time this occurred (intraday high of 2118 followed by a next-day low of 2085). From May 12 onwards -through today (May 21)- we have gone 7 days. During which time the markets on a later date have still not fallen at least 1% from (any of) those 7 intraday highs.
A streak of 7 trading days is also a lengthy period of time. During the past 2 years (>500 trading days), this streak of >7 days without an eventual decline of at least 1% has only happened 3 times. Otherwise there have only been a handful of times when the streak was capped at just 3 or fewer days, again without this 1% decline. This all implies an exceptionally, low probability event.
The recent market data, which clearly may not repeat in the same way going forward (only dim-wits construct alarming predictions: here, here, here), simply indicates to us that our current low volatility streak is rare. Generally occurring only once annually, and when it does it typically only lasts about 9 days: so about another day or two to go!
Much more likely though (say a >2/3 probability or so), we are just not in a record streak, and that on a future date we would have at least enough of a correction to contract this current market pattern back to its historic norms mentioned earlier. If this were true, then the May 15 intraday high of 2124 must see an eventual price of at least 1% less, so that the historic streak is capped at the otherwise typical 3 days noted above. Also, a 1% decline from this S&P of 2124 equals 2103 (a level that is a 1.3% discount from today's close of 2131). Stay tuned!
Incidentally, for those interested in an auxiliary topic of market-drop convolution, please refer to the treatment given in this article, or search for other related topics on the blog.
No comments:
Post a Comment