Friday, February 12, 2016

Retail Sales Up … Michigan Sentiment Down… Stock Market Analysis

RETAIL SALES (MarketWatch)
“Retail sales rose 0.2% in January, as consumers boosted purchases of new cars as well as groceries and shopped more online…Sales in December were sharply revised higher to show a 0.2% gain instead of a 0.1% decline…” Story at…
“Consumer sentiment declined in February to a four-month low as declining stock prices and weaker global conditions weighed on Americans’ views of the economy. The University of Michigan’s preliminary index decreased to 90.7 from 92 in January…” Story at…
-Friday, the S&P 500 was up about 2% to 1865 at the close.
-VIX was fell about 10% to 25.4.
-The yield on the 10-year Treasury rose to 1.75%.
There were cross currents Friday.  There was a significant improvement in new-highs vs new-lows even though new-lows still outpaced new-highs.  The size of the improvement by itself does not signal a bottom, but it might signal the short-term bounce I have been expecting.  On a negative side, the size of the up-move was statistically significant and that means that the price-volume move UP exceeded my statistical parameters and, in about 60% of the time, that leads to a down-day the next day (Tuesday).
Breadth is now oversold, but that will clear quickly if there is another up day on Tuesday so I don’t think it is important. (RSI is not oversold.)
There has not been a buy signal in internals or indicators since the sell signal last December.
The smoothed version of the Money Trend remains clearly down suggesting the overall market trend.
I took a very small, short-position to go opposite the big up-day, but it wouldn’t surprise me to see that trade go sour since the S&P 500 could easily move up to the 1920 region.  It could just as easily drop back a bit first, so my trade is probably just a toss-up.  I would not have moved at all today, but 2%-up-days are usually followed by a down-day.  We’ll see.
(I am getting data from various sites. Some of the numbers are subject to minor revision so the previous day’s numbers may be slightly different than reported yesterday.)
The 10-day moving average of the percentage of stocks advancing (NYSE) is 42.3% Friday vs. 43.3% Thursday. (A number below 50% is usually BAD news for the markets. On a longer term, the 150-day moving average of advancing stocks remained 48.5%. A value below 50% indicates a down trend. The McClellan Oscillator (a Breadth measure) improved, but remained negative on the day.
New-lows outpaced New-highs. The spread (new-highs minus new-lows) was minus-116 Friday. (It was -672 Thursday.)   The 10-day moving average of the change in spread rose to minus-14. In other words, over the last 10-days, on average; the spread has DECREASED by 14 each day. Market Internals (based on 10-dMA) remained negative on the markets.

Market Internals are a decent trend-following analysis of current market action, but should not be used alone for short term trading. They are usually right, but they are often late.  They are most useful when they diverge from the Index.  In 2014, using these internals alone would have made a 9% return vs. 13% for the S&P 500 (in on Positive, out on Negative – no shorting).  Of course, few trend-following systems will do well in an extreme low-volatility, nearly straight-up year like 2014.
Friday, Volume and VIX were negative; Price & Sentiment indicators were neutral. The long-term NTSM indicator is SELL. At this point a sell indication just reiterates that market conditions are poor.  The first SELL signal of this cycle was 18 Dec 2015 and there has not been a BUY signal since.

On 30 Dec I reduced my invested position in my retirement account to 30% invested in stocks thru an S&P 500 Index fund (“C”-fund in the TSP). Friday, 15 Jan I reduced stock allocation to zero in long-term accounts. That leaves 100% invested in cash yielding about 2%.  Short-term bonds would be OK too.
The S&P 500 peaked in Mid-May and has not been able to break higher in the past 9-months. That looks like a top to me. See “Why the Bull Market May be Dead” in my 14 December blog at…
Even if that is true, there could still be a rally for 2 or 3-months.