Short-term Outlook:  Neutral  From Feb 26, 1107 SPX

Equity Indicators - Updates and Extremes

NYSE Up Volume Ratio

We've spent some time going over the various streaks that have popped up over the past week or so, in the VIX, Russell 2000 and yesterday the S&P 500.

That kind of persistent uptrend will tend to generate some extremes, and we're seeing them among our indicators at an increasing pace.  One very notable development is in the Up Volume Ratio for the NYSE.

The ratio simply looks at the volume going into stocks that are up and stocks that are down each day, and calculates the percentage of volume that went into the advancing stocks.  So a reading of 65%, for example, would mean that 65% of all volume that day went into rising stocks.

On the site, we show this data via three moving averages, short-term (5-day), intermediate-term (10-day) and long-term (21-day).  As of yesterday, all three moved above 65%.

Extreme?  You bet...it's the first time all three were that overbought at the same time in nearly 20 years.

If we restrict our search to only those times when the S&P 500 was near (within 1%) of a new 52-week high at the time, we get about a dozen instances since 1950.  The vast majority of those were found in the 1950's, with only four since then.

Looking at the latest four instances, the market did OK afterward, but nothing too exciting either way.  Six months later, the S&P was still within 10% of where it was when it reached such a phenomenal overbought level.

In '63, the S&P went absolutely nowhere for three months before resuming its long-term rise.  In '76, it rallied hard for a month...and then essentially didn't do anything for a year.  In '78, stocks chopped up and down in a 4% range for two months, then dropped hard for a month where it formed a major low.  In '91, the S&P again got stuck in a tightly ranged market with modest upside bias for nine months.

STEM.MR Model

One real oddity that occurred yesterday was that despite equities holding up well during the day and closing at a new one-year high, our ultra short-term STEM.MR Model actually closed near oversold territory.  Now that's weird.

Going back to 2002, there have only been four other days when we've seen such a thing.  The table below shows the returns in the S&P going forward.

 

Date	1 Day Later	1 Week Later	1 Month Later
12/04/03	-0.7%	0.3%	4.6%
07/14/05	-0.1%	-0.2%	0.1%
07/22/05	-0.3%	0.2%	-0.9%
10/10/06	-0.1%	0.8%	2.7%
Average	-0.3%	0.3%	1.6%

All four closed lower the next day, though not by any more than -1%.  And their recoveries were quick, showing mostly positive returns in the day(s) following.

The instances in 2003 and 2006 led to steady positive returns over the next few months, but the two from July 2005 saw the S&P form a minor intermediate-term peak a little over a month later.