Mojena Market Timing

 Front Page      Timing Model      Reality Check      FAQ      Downloads      About      Mea culpa?

December 28, 2014

Model at 61.0

SELL Signal on December 14, 2014*

Strategy

Current Position

2014 Returns

Cash

Money Market (T-Bills)

+0.0%

Buy and Hold

100% S&P 500

+13.7%

Standard Timing

Money Market (T-Bills)

+9.8%

Aggressive Timing

100% short S&P 500

+7.3%

 

 

*The timing model issues buy and sell signals based on a mathematical/statistical score that ranges between 0 and 100:  Sell 39 or below; buy 76 or above. The standard and aggressive strategies determine the trades when timing signals are given.  Signal date is Sunday; trades based on next-day closing price.  Returns include reinvested dividends.

Changes (no dividends) in indexes since SELL signal… DJI: +3.7%    S&P 500: +3.5%    Nasdaq: +2.8% 

Date

Model

Signal

S&P

Cash

B&H

Std

Agg

<< YTD returns, including dividends

Dec 28

61.0

Sell

2089

+0.0%

+15.3%

+9.8%

+5.7%

Rally continues second week as Dow and S&P set new record highs.  The primary uptrend is now reconfirmed.  Model improves, but stays on sell signal.  This could very well be a faulty sell that leads to regretted gains at the next buy.  Or at the very least it was an early sell by at least two weeks.  This model’s historical record shows that 38% of the sell signals (11 of 29) resulted in missed gains averaging 3%, ranging from 0 to 6%.  The 62% successful sells avoided losses of 12% on average, ranging from 1 to 48%.  These asymmetric results suggest that it’s wiser to follow sell signals than not.

The model for 2015 will include this year’s data as part of its genetics.  It should be operational within the next couple of weeks.  Meanwhile we continue with the 2014 model.

Here’s wishing you a happy, healthy, and prosperous 2015

Dec 21

48.1

Sell

2071

+0.0%

+14.3%

+9.8%

+6.6%

Massive two-day rally catapults the S&P from a 5% pullback to just under its all-time high the previous week.  The model doesn’t buy this major bounce, staying on its sell signal.  Ok, so the market was oversold, the Fed’s “patience” remark on future interest rate increases lit the fire, concerns eased over too-low oil prices that would damage currencies and emerging markets, and short covering all probably contributed to the stellar comeback.  Still, the strong reaction was surprising, if not shocking, yet the model stayed firmly within its sell signal.  Why?  Positive turns for two key technical indicators (see Aug 24 for relative strength and golden cross) were not enough to more than compensate for continued deterioration in market confusion, as measured by its Hi/Lo index (see Dec 7).  Moreover, risk as measured by the VIX Index, while improving, remains elevated, a negative for the model.  Finally, the model is less sensitive near its extremes, more reluctant to move dramatically, the existing condition last week.  Its current score, however, is in a range that’s more sensitive to change.

The model could be wrong here, especially given that its portfolios now lag buy and hold.  But then, it’s only been one week  the final judgment will be rendered when it’s over at the next buy signal.  It ain’t over ‘til it’s over.  Or… maybe not even then.  Looking at the historical record for this year’s (2014) model, the turbulent bear markets in 2000-2001, 2002, 2007-2008, and 2009 precipitated a series of sell/buy/sell signals, not all winners, yet in the end the key sell signals avoided the major damages caused by those bears. Meanwhile, we (the model and I) hang in there during the current sell signal. 

Have a very Merry Christmas and Happy Holiday week

Dec 14

16.2

Sell

2002

+0.0%

+10.5%

+10.5%

+12.2%

The model has issued a sell signal, effective at the close on Monday, December 15.  Accordingly, the model’s standard timing portfolio switches 100% into a money market fund based on T-Bills and the aggressive timing portfolio shorts the S&P 500 by 100%.  Over the buy signal, index gains (not counting dividends) for the Dow, S&P, and Nasdaq are respectively 12%, 17%, and 22% Final numbers were based on Monday’s close, to conform with the standard practice of mutual funds marking to market at the end of the trading day.

A “pure” implementation of the standard strategy during a sell signal is to shift all stock funds into a money market fund, as done in the table above.  This “all-or-nothing” buy-sell strategy is uncommon in practice, as most portfolios would be diversified across several stock classes while invested and would be reluctant to sell all holdings; it’s used  here to give a proper comparison to buying and holding the S&P 500.  In practice a portfolio that follows the model’s sell signals could either sell all stock holdings or sell a portion, leaving other holdings that might reasonably weather a serious market downturn… or as a hedge against a faulty sell signal.  In other words, do whatever you’re comfortable with given the information that the model is now on a sell signal… and to what extent you have confidence in the model.  I will be all out over the coming week on this clearly unambiguous sell signal, except for a small position relative to my portfolio in a targeted ETF that’s somewhat illiquid (CUBA). 

Portfolios that mimic the aggressive strategy would shift all stock funds to a mutual fund that shorts (places a bet against) the S&P 500, such as Rydex Inverse S&P 500 (RYURX) or ProFunds Bear (BRPIX).  Or shift the entire portfolio to the ETF SH, which also shorts the index 100% (-1x) on a daily basis.  In other words, given a 2% loss for the S&P 500 by the end of the trading day, the short should give a 2% gain.  The other edge works against us: a 2% index gain would be a 2% loss on the short.  Unfortunately, this position deteriorates over periods of more than one day based on mathematical volatility. See this CAUTION.  Please note:  The pure aggressive strategy is very risky and should be practiced, if at all, with a small portion of the overall portfolio, as I do with less than10%. 

See a more detailed explanation of these options in the three FAQs starting with this one. 

The model’s dramatic plunge to a score well below its sell trigger, with the pullback just 3.5% under the S&P’s high just one week ago, tells me that we could have a more serious decline over the coming months, unlike the last five-week sell signal in August, 2013, which was issued with a score barely below its sell trigger and was accompanied by just a 5% pullback.  That sell signal gave up regret gains of 5%.  But then… if we had ignored the model’s sell signal in Jun 2008 we would have lost about 26% of our portfolio (based on the S&P 500) by the buy signal in Oct of that year.  This sell signal looks more like the one in 2008 than that in 2013. 

Dec 7

77.1

Buy

2075

+0.0%

+14.4%

+14.4%

+18.5%

Another week, another set of ending highs.  This past week the financial media resurrected the Hindenburg Omen, a technical indicator that had just issued a signal that portends disaster, the financial correspondent of the physical calamity in 1937 of that famous German zeppelin.  While the Omen has done a good job of predicting market declines, it’s also done a poor job by generating many false positives.  The model does use a version of the Fosback Hi/Lo Logic Index, mentioned here within the context of the Omen.  This Hi/Lo indicator gave a false sell in June last year, which was followed by a short and uneventful 5% decline.  The model’s “omen” based on the Hi/Lo index has deteriorated over the past several weeks, contributing to the model’s declining score, despite the S&P advance.  The model’s score is now in a range that’s more sensitive to events.  The takeaway here is that the complexity of the market calls for a model that looks at more than one or two or three indicators.  The model uses 16 indicators across four influential categories: technical, monetary, sentiment, and fundamental.

Nov 30

86.3

Buy

2068

+0.0%

+14.0%

+14.0%

+17.9%

Quiet Thanksgiving week ekes out new record highs, for the Dow and S&P.  The Nasdaq?  It’s now just 5% below its all-time high in March, 2000, a more than14 year slog that would surely test the patience of a buy and holder for that index.  On the other hand, a perfect timer buying at its subsequent bottom would see an index gain (without dividends) of 330%, compared to 172% for the Dow and 206% for the S&P.  These results show the value of including elements of the tech-heavy Nasdaq within a portfolio.  Greater volatility yes, along with greater returns.  That is, as long as a reasonably good timing strategy is used, to tone down volatility and to mostly avoid the devastating bears that characterize so-called high-Beta investments.  The Nasdaq cleavered 78% of its value over a 31 month plunge into its 2002 bottom.  During buy signals my own portfolio includes at most a 15% allocation of a large-capitalization and non-financial subset of this index, the Nasdaq 100 (QQQ) and up to 10% in a biotechnology subset (IBB for larger biotechs or XBI for smaller biotechs that include holdings within and outside the Nasdaq).  Keep in mind… these are volatile ETFs that should be used with caution and patience. 

Nov 23

89.2

Buy

2064

+0.0%

+13.7%

+13.7%

+17.5%

Dow and S&P end week at new records.  We got stock market help from Chinese and European central banks, based on the assumption that stimulative monetary policies will boost weak growth in their respective economies, allaying fears of a global slowdown.  Other positives include sustained low interest rates, a merger and acquisitions boom, continued stock buybacks, strong purchases of U.S. stocks by overseas buyers, and the favorable Holiday season. 

What’s not to like?  Let’s see… acrimony grows in Washington following the President’s possibly illegal and apparently misleading executive order on immigration.  We absolutely need immigration reform, both for economic and humanitarian reasons, the latter including a path to legality while ensuring strengthened border security.  Maybe this will encourage legislation to replace the executive order, a compromise in needed immigration reform by Congress in 2015, generally viewed as an economic positive by most economists and the CBO.  Examples? Let’s double the specialty H-1B visas and fix the guest-worker program.  And let’s have other important pro-growth and business-friendly bills move forward as well.  If not …? 

We’re up 11% from the closing low just five weeks ago.  Is this an early Santa rally with little left through year end?  The model places odds of 9 to 1 that the primary uptrend remains in place.  That settles my bet.

Let’s put aside our concerns and enjoy a Happy Thanksgiving.

Nov 16

89.2

Buy

2040

+0.0%

+12.3%

+12.3%

+15.5%

Quiet week as Dow & “500” stretch record highs.  A common bull-market pattern shows a range-bound stock market following a surge, then a renewed extension of the uptrend, so we’ll see should the market trade within a small sideways range over the coming handful of weeks.  The market should respond positively to the current favorable season, the lower oil prices spurring consumer demand (but hurting the S&P’s oil sector), and what now looks like passage of the Keystone XL pipeline by the House and Senate during the current lame-duck Congress.  Possible negatives leading to another pullback? Too strong a dollar seriously hurts the oil sector and exports, a likely Presidential veto of the pipeline, executive actions on immigration that “poison the well” on cooperation between Congress and the President, and a serious and sudden geopolitical event such as an overt Russian invasion of eastern Ukraine.  The drama continues and the model monitors.

Nov 9

89.6

Buy

2032

+0.0%

+11.9%

+11.9%

+14.8%

Dow and S&P 500 add to record run.  Republicans sweep the election by controlling the Senate, extending the House, and adding to Governorships and State Congresses.  The implications of this political wave for the stock market?  Not easy to say.  For sure the blockade in the Senate will end and the President will find much more legislation on his desk, including bipartisan bills on energy (Keystone approval, gas exports, maybe some eased carbon emission rules), trade agreements (Asia, Europe), unpopular aspects of the ACA (elimination of certain Obamacare taxes, more choice), some taxation and regulatory changes (especially for businesses), and piecemeal immigration changes (such as revised and expanded work permits for guest workers, visas and Green Cards for skilled and highly-educated workers, unless the President goes ahead with executive actions that undermine cooperation).  In time these changes should improve the job market and add to GDP growth, actions which should compensate for the eventual tightening of Fed interest rates and weakness overseas.  In short, Washington’s legislative log jam should ease, especially for business-friendly changes.  The keyword is should.  I’m optimistic should a spirit of cooperation dominate proposed legislation that the parties and President see as common ground.  And compromise when not… on, say, immigration, health care, infrastructure, tax reform, regulations on small banks and small businesses, defense, minimum wage, welfare reform.  The country voted for change, done with hope.  At the very least legislation will be more transparent as it reaches the President’s desk via the traditional route of committees, conferences and floor votes, procedures that have mostly been bypassed by the Senate’s leadership freeze. 

Moreover, we do have some stars that are aligning for the market:  We’re entering a favorable annual seasonal period and the stock market has averaged double-digit yearly returns following midterm elections since 1946.  Speculations aside, the model listens, an electronic fly on the wall, and passes its own legislation.

Nov 2

91.2

Buy

2018

+0.0%

+11.1%

+11.1%

+13.7%

Markets extend surge, taking Dow and S&P to new all-time records.  The primary uptrend dating back to August 2011 is confirmed, vindicating the model’s view at the recent pullback low (see Oct 12 & 19 posts).  The jump from the intraday low in mid-October is 10.8%; from the closing low it’s 8.4%.  By definition, the bull market that began in March 2009 remains intact in its sixth year, with a near-tripling (+198%) for the S&P 500 (ignoring dividends).  Tack on another 24 percentage points (+222%) to account for reinvested dividends. 

To reinforce a point made in several places within these pages: The main benefit of a good timing system is to capture most of the gains during long (months to years) uptrends and avoid most of the losses during long downtrends, while reducing one or more measures of risk.  An ancillary but very important benefit is the available discipline of a strategy that avoids the emotional mistake of buying high (when fear is low) and selling low (when fear is high), especially for short yet decisive market swings.  Several studies and many articles call attention to this damaging behavior by retail investors and, yes, far too many professionals.  For example, see this and this.  See also the Jun 22 post. 

Oct 26

83.3

Buy

1965

+0.0%

+8.1%

+8.1%

+9.2%

Markets pop for the week as investors jump back in: 2.6% for the Dow, 4.1% for the S&P, 5.3% for the Nasdaq.  The closing pullback now stands at 2.3%, after reaching its current closing low of 7.4% last week with an S&P 500 closing print of 1862 and intraday low of 1821 (9.5%).  The jury is still out regarding the book for this pullback, so long as the S&P remains below its record high of 2011 set last month.  If this pullback has seen its low, the trading week ending October 17 presented a nice opportunity for buying stocks at discounts.  Still, more openings will show, given continued volatility and providing the model stays on its buy signal.

Oct 19

71.8

Buy

1887

+0.0%

+3.8%

+3.8%

+2.8%

Wild week marked by panicky day on Wednesday, as the Dow plunged 450 points and the “500” hit a 9.5% pullback from its September intraday high, very near a 10% intraday correction, although both indexes ended up with far less losses by the close.  On a closing basis, the S&P marked a 7.4% pullback, a routine pullback (5 to 10%) that averages about 3 per year in about half the years.  Earlier this year we had a 6% dip.  Given that we’ve had a routine pullback, it becomes a moderate correction (10-15%) with a frequency of about 25%, severe correction (15-20%) 17% of the time, and bear market (20% or more) the remaining 8%.  The model says that this should stay a routine pullback, with a likelihood of 72%, well above the expected 50% given the historical record.  The median break (half above, half below) is 8.2%.  It’s not likely that “the sky is falling.”

We’re in just about the only business in which discounts are greeted with alarm.

RRob Arnott, Chairman, Research Affiliates

Wednesday’s intraday low might have been a capitulation of sorts, meaning that we are likely near the lows.  Still, we could easily test the lows over the coming weeks, especially if (pick one or more) corporate earnings disappoint, more Ebola cases appear in the industrialized economies, geopolitics worsen appreciably, and the global economy weakens more than expected.  Or not.  Expect volatility at least through the U.S. election in early November, another unknown that will surely affect the stock market.  More on that political factor in an upcoming post.  Meanwhile, let’s listen to what the model says and hope it’s right.

Oct 12

73.6

Buy

1906

+0.0%

+4.8%

+4.8%

+4.4%

Surprise for sure.  I had been expecting a sell signal, given the status of the model last week and the additional market decline this week.  Instead, the model not only stayed on its buy signal but counter-intuitively strengthened as well.  While some of its technical indicators weakened further, two important contrary technicals turned positive, one dealing with extreme lows for the week and the other a relationship between the Dow Jones Industrials and Transports.  The last time these two signaled simultaneously was in August, 2011, two weeks before the formation of a climactic week-ending bottom and eight weeks before the eventual short-lived bottom for that year, a bottom that was near the August low.  From April to the bottom in October the S&P 500 put in a severe 19% correction over 16 weeks.  The model missed that correction (its worse historical mistake), incorrectly and stubbornly staying on its buy signal.  Yet, from that bottom the “500” surged 12% over the next three months into the end of the year, ending the year with an overall gain of 2%, as did the model, and preserving the ongoing bull market to this day. 

That year was a difficult one for both the market and the timing model.  This year could be as well.  The  market is currently oversold, although given the fear out there we could easily have more big down days.  The market swooned 7% on the Monday following the simultaneous contrary signals in 2011, very near the eventual bottom in early October for that correction, reinforcing October’s fame for market bottoms.   The historical record shows that these simultaneous contrary signals are associated with volatile markets.

So, where do we stand?  The model is saying that the primary uptrend remains intact, with about 74% “certainty.” And by extension the bull market remains in play.  It’s the 26% that worries me.  The pullback now stands at 5% over three weeks, not unusual; a new primary downtrend would require a loss of 8% or more from the recent week-ending high over at least 8 weeks. The model does not anticipate this happening, but it sets odds of 1 in 4 that it could happen.  Moreover, I now don’t see the model issuing a sell signal anytime soon.  Meaning that, should the pullback continue or even enter correction territory (10% or more loss), there could be buying opportunities for under-invested portfolios, especially on a big down day, such as the 7% Monday in 2011.  Note that corrections over less than eight weeks fall under the model’s radar, by design, to preserve the model’s overall historical performance while maintaining its intermediate to long-term outlook without excessive switching.  The model says stay the course for those fully invested, if you can deal with any downside volatility.  October bottoms, upcoming earnings reports, and a favorable season starting in November might save the day.

Oct 5

43.5

Buy

1968

+0.0%

+8.1%

+8.1%

+9.6%

Volatility continues for wobbly market; model follows suit as it settles about 4 points above its sell trigger.  The “500” is only about 2% below its all-time high from just 12 trading days ago.  Intraday it dropped a bit over 4% on Thursday.  October, 2007 shows a similar confluence of pullback percent and score for this current model’s historical data.  At that time the model barely issued a sell signal with the S&P about 4% below its recent high.  A couple of switchback signals followed before the model settled on a 71 week sell signal that mostly avoided the 57% devastation during the 2007-2009 twin bear markets from the financial collapse.   Still, there have been many instances in the historical record when the model rebounded from near-sell scores.

History doesn't repeat itself, but it does rhyme.

Mark Twain

A sell signal is now possible week to week, especially given the uncertain consequences of a couple of fat tails out there: The Hong Kong demonstrations, how the Chinese government reacts, and what effect if any it will have on their economy, and by extension other interconnected economies; the Ebola crisis, which could turn into a growing global pandemic, although far from likely.  Of these two, the latter is considerably more important, both from humanitarian and global economic perspectives.  The increasingly soothing statements from our government are paradoxically worrisome.  (I’m thinking of the calming Fed statements as the 2008 housing and financial crises initially unfolded, as one of several recent examples, not just in finance.)

With America’s leadership… this epidemic will also be stopped.

Our healthcare infrastructure … is well equipped to stop Ebola in its tracks.

Senior White House Officials

Yes, we have excellent protocols and facilities for containing its extent, and past outbreaks have been stopped, but then today’s global travel can exponentially spread a virus.  And the mistakes made in Texas don’t exactly give me confidence that it will be contained there and elsewhere.  Expect more Ebola cases in the U.S. and initial cases in Europe and elsewhere.  A global pandemic is devastating for economies.  Traders are worried.  It actually might not take a pandemic to tip many economies into recession; multiple serious outbreaks might do it, thereby affecting shaky stock markets.  Let’s hope this pessimism is unwarranted.

Sep 28

62.3

Buy

1983

+0.0%

+8.9%

+8.9%

+10.8%

Volatility is back, with the Dow printing triple-digit daily losses and gains each day over the past week.  Yet, the “500” eased back just 1.4% from its all-time high the previous week, during the last full week of trading in September, considered the weakest week of the year for stocks, based on historical averages.  The continuing push and pull between bulls and bears (see last week) adds to volatility even though the trend line for the index is upwardly sloping for the year.  The model includes a measure of variability about the trend line that is up about 33% since the beginning of the year.  (It’s an exponentially smoothed moving variance, for those of you familiar with this statistical metric.)

Likewise, the model itself has been volatile lately, as several of its technical, monetary, and sentiment indicators fluctuate about triggers.  This means that the model bears watching for a possible upcoming change in signal.  Meanwhile, downside volatility is not all bad, as it shakes out the weak players (to return later) and offers opportunities for adding to stock positions, for underinvested portfolios and while the model remains on its buy signal.

Sep 21

83.3

Buy

2010

+0.0%

+10.4%

+10.4%

+13.2%

The Fed was kind, the Scots voted NO, and the Dow and S&P are back in record territory, the latter at 2011 and nearly triple its bear market low back in March, 2009.  

The Bears

The Bulls

High valuations (by some measures) mean the bull’s end is near

See Jul 13

OK valuations (by some measures) mean the bull’s end is not near

See Jul 13

New highs in an aging bull market mean we’re close to the end

New highs confirm bull markets

We’re in a bubble (lots of people say so)

We’re not in a bubble (lots of people say so), especially given plenty of cash on the sidelines and fairly high skepticism

Interest rates are going up, a negative for profits and the economy

Dovish Fed will maintain low interest rates while the economy struggles and, at any rate, an improving economy will compensate

This market is delusional given stuttering recovery in the U.S. and stagnant economies in Europe

Stock markets like Goldilocks economies with low inflation and good profits

Seasonals into October are scary, given  the historical record

See Aug 17

Seasonals just describe averages, not what will happen in any particular year

See Aug 17

Unsettling geopolitical events are ripe for negative fat-tail events

See Aug 31

Maybe, but most geopolitical events have little impact on global economies

See Aug 31

So, who’s right?  Probably both, depending on time frame and evolving events.  It’s confusing, to say the least, so I let the model sort it all out.  The primary uptrend and buy signal remain in place, as is the strategy of buying dips for underinvested portfolios. 

Sep 14

77.5

Buy

1986

+0.0%

+9.0%

+9.0%

+11.1%

The “500” eases back 1% from last week’s high; the model responds in more than kind, as the death cross replaces the golden cross (see Aug 24).  We could have some additional volatility in the coming week: as investors parse the Fed’s policy announcement on Wednesday for clues as to interest rate intentions (increases sooner than next summer?); the results come out on Friday from the Scottish independence vote, an important financial disruption should the YES vote win, given that Great Britain is the sixth largest world economy by GDP and a member of the European Union; and the Alibaba IPO, the largest in history, debuts trading on Friday, a likely positive event for the market (will it compensate should the others be negative events?).

Competition Is for Losers

If you want to create and capture lasting value, look to build a monopoly.

Peter Thiel

Sep 7

90.9

Buy

2008

+0.0%

+10.2%

+10.2%

+12.9%

S&P 500 ends week by extending record close to 2008.  Beginning season of infamous declines into October starts with a whimper.  The model’s current strength puts it in a good frame of mind to withstand expected pullbacks during this month and next.  Its technical indicators are solid once more, as are the monetary indicators.  Sentiment is neutral, just enough fear to balance out optimism; nowhere near extremes in either direction for turning-point concerns.  Fundamental indicators such as price to earnings and dividend yields look okay.  Geopolitics remains a wild card, as always, although more wildcards stack the deck these days.  The primary uptrend is solidly in place, as is the operating strategy during a buy signal: Underinvested portfolios should consider buying the dips.

Aug 31

91.0

Buy

2003

+0.0%

+9.9%

+9.9%

+12.6%

Another week, another record high at 2003 for the “500.”  And this despite a Russian incursion invasion in eastern Ukraine, another red line crossed.  Why would the stock market worry?  Ramped up sanctions primarily between Russia and Europe might tip an economically wobbly Europe into recession, with implications for economic health on this side of the Atlantic and elsewhere, such as China.  Apparently, reasonably good stateside economic news, robust earnings, the kind Fed, few investment options showing return other than equities, and buy-the-dip successes during this bull market are, for now, trumping geopolitical heat.  Investors are either complacent regarding the potential consequences from global conflicts or figure that continued events and new sanctions would not be severe enough to meaningfully alter the positives. 

Whatever, the market keeps trucking along and the model keeps an eye on things.  Speaking of things, here’s a scary known unknown Black Swan : The catastrophic consequences of a widespread electromagnetic pulse (EMP).  Check it out here.  Ok, while this probability tail might be fatter than we think, it’s probably not a good strategy to base investments on rare doomsday scenarios that encourage us to turn all investments into physical gold and join the survivalists in the wilds.  So, maybe we make some appropriate preps, but primarily follow the model and do buy the dips for underinvested portfolios while the buy signal is active.  And dips are definitely common in the stock market’s worse month, now upon us.

Aug 24

89.7

Buy

1988

+0.0%

+9.0%

+9.0%

+11.3%

Primary uptrend reconfirmed as the “500” marks 1992 all-time high this past Thursday.  The mild 4% pullback is now history, joining the moderate 6% pullback earlier in the year, as temporary retreats from this relentless five-year bull market.  The model jumped 12 points this week and 32 points over two weeks, pumped by improved technicals, especially from positive turns by its versions of relative strength and the golden cross (the good cross that reverses the death cross). 

We could see some volatility in coming weeks, as the “dog days of summer” end with thin trading and especially should 20th and 7th century-style geopolitics in Ukraine and Syria/Iraq heat up further within the persistent and ageless realities of geopolitics in the 21st century (naïve to think otherwise?).  Existential angst aside from over-thinking the market, the S&P 500 should continue its ascent through autumn, barring any sudden or earlier than expected rate jump by the Fed and given the large amount of unproductive cash on the sidelines that’s increasingly committed to equities by investors chasing returns with few options other than the stock market.

Aug 17

77.3

Buy

1955

+0.0%

+7.1%

+7.1%

+8.5%

Market and model cautiously advance onto more solid footings as bears and bulls continue their tug of war: earnings favor the bulls, geopolitics favor the bears, and the Fed could jump from one side to the other, previously favoring the bulls but hedging with the bears over the coming months.  And is the Russian bear poised to disrupt the contest once again? 

And will the season have an effect?  September is around the corner, the worst performing month for the stock market, on average.  A common pattern is an interim top in July or August, a selloff in September into October, and a bottom as a springboard into the generally favorable holiday season.  A common pattern, however, is far from a certain pattern.  Research with the model shows no advantage to trading the season.  Yes, a negative seasonal effect in a particular year, to the extent it influences the model’s technical indicators, could tip the model into a sell signal when it’s already close to a sell signal, but not otherwise.

A new pdf file for the technically inclined (Lecture file from my MBA classes)

Financial Math 101

Aug 10

57.5

Buy

1932

+0.0%

+5.8%

+5.8%

+6.6%

Wobbly market edges up for the week as the push and pull of geopolitical and economic events play out.  The model clings to the 50s, while remaining vulnerable to a sell signal in the coming week.  As a group the model’s technical indicators have weakened over the past couple of weeks, with several, such as an intermediate moving trend, remaining positive enough, while others, such as its version of the death cross, turning negative.  A contrarian sentiment indicator accounted for most of the model’s change this week, by taking on a more positive tone in reaction to a rise in investors’ pessimism. 

The pullback last week fell just shy of 4% and now stands at just under 3%, so the question remains:  Is the primary uptrend still in place, or is the pullback part of a new primary downtrend?  The model favors the former and remains on its buy signal.  See this chart for a view of trends and how an oversold S&P 500 has bounced off the lower Bollinger Band over the past six months.  A more sustained breakdown below this band would likely trigger a new sell signal.

Aug 3

51.2

Buy

1925

+0.0%

+5.4%

+5.4%

+6.1%

Angst turns to selling as model and unnerved market swoon.  Contributing events? Europe:  Weak economy, deflationary fears as in Japan, and more anticipated consequences over Russian sanctions.  Argentina: Sovereign bond default leading to possible contagion while interrelated creditors and bond holders are hit and emerging markets are put at risk, including peripheral Europe, as in Portugal.  Fed: Will they raise rates sooner than expected based on a strengthening economy and incipient inflation?

The market had weathered lots of bad news up until now, but nevertheless kept rising.  Complacency eventually leads to volatility, yet the current 3% pullback is garden variety.  We could easily see a total 6% or so pullback during the upcoming week, as earlier in the year.  Still, earnings are good, the labor market appears stabilized (although lots of part-time and discouraged workers), the economy is growing and an even better second half is expected, geopolitical events might tone down, the Fed will assuage and just might get it right over the coming year, and there’s plenty of unproductive cash on the sidelines. 

The model took an unexpected plunge this past week based on weakened technicals and is now very sensitive to further market deterioration.  A sell signal is a real possibility by this time next week, should the score drop to 39 or less.  At this week’s score of 51 the model calculates odds of about 50/50 that the current 3% pullback is consistent with a primary downtrend, a loss of 8% or more over eight weeks or more, based on Friday closings.  But then… the market could very well bounce back by week’s end  should investors buy the dips, as they have during this bull run.  So we’ll see.

Jul 27

92.7

Buy

1978

+0.0%

+8.3%

+8.3%

+10.5%

The “500” ekes out new 1988 record before easing back on Friday over the usual current angst: geopolitical events, the global economy, the U.S. economy, the Fed, corporate earnings, the weather, valuations, bubble, what to eat and what not to eat… you name it we’ve got it.

A reader recently asked “Your model has 4 indicator categories. Do you mind sharing the approximate weight of each category?”  This is a great question that actually never occurred to me, let alone the answer.  Unfortunately, this simple question requires a not-so-simple answer regarding the "weight" of each category  (technical, monetary, sentiment, fundamental).  The answer would be easy if the model were linear with point scores assigned to each indicator within a category; the category weights would then be the sum of indicator points within each category. This is far from the case. 

But... maybe this would help.  Of the 16 indicators, 10 are tech, 3 monetary, 2 sentiment, and 1 fundamental.  But even this description is slightly misleading, because several include elements from two categories, e.g., fundamental mixed with tech or monetary, monetary mixed with tech.  Moreover, some indicators are composites of underlying indicators.  For example, one monetary indicator is a composite of six monetary measures.  We can safely say, though, that technical indicators dominate.

The answer, without further explanation, is that technical indicators are the most significant, followed in order of importance by monetary, sentiment, and fundamental. Technical indicators likely predominate because these are mostly driven by the psychology of supply/demand (fear/greed, sell/buy) and include "the anticipation of others' anticipation," as the reader aptly put it.  In the end it's the composite score that matters relative to the buy/sell bands.  The current score of nearly 93 is well above the sell trigger at 39, so we remain on a buy signal.

Jul 20

93.8

Buy

1978

+0.0%

+8.2%

+8.2%

+10.5%

Market swoons on Thursday over commercial jetliner’s downing by missile, recovers on Friday, and ends the week higher.  Pullback now at 0.4%, reaching 1.4% at the week’s low.

Geopolitics and global economic implications are back on the table with new Russian sanctions from the U.S. and the downing of a Malaysian jetliner by a Russian-made missile.  The jetliner calamity is refocusing global attention on the Ukrainian conflict given the international scope of this tragic event:  An American-made aircraft flown by an Asian airline company from a Eurozone country flying over Eastern Europe with 298 passengers and crew from ten countries, mostly Dutch, Malaysian, and Australian.  That’s global reach that will put more heat on the Russians and add more uncertainty to unfolding events in that conflict.  And then we have military engagements in Syria, Iraq, Israel/Gaza, not to mention belligerence between China and its neighbors in the South China Sea.  As if that’s not enough, new and continued sanctions on Russia will likely lead to retaliations, putting global economic recovery at risk.

Overall, the stock market has handled all of this rather well, but then this kind of global uncertainty unsettles traders and increase market volatility, as seen by Thursday’s selloff.  And there’s always the possibility of more escalation and even a Black Swan event that leads to a significant correction or bear market.  Or… the market shrugs it all off, believing global growth will remain on the right track, which appears to be the case thus far.  The model considers, as we do.

Jul 13

94.0

Buy

1968

+0.0%

+7.6%

+7.6%

+9.6%

Market eases back about 1%, the catalyst this week renewed concern over Europe’s sovereign debt crisis, with Portugal in the crosshairs.  The issue never really went away, just simmering under the surface.  We will likely see more reactions in the future.  It will take time to fix.  Search for “Europe” in the 2011 , 2012, and 2013 archives for more on this topic.

The minor pullback this past week brought out the media bears once more, with talk of a correction (down 10%+) or even a bear market (down 20%+).  Why?  (a) A correction is overdue, (b) the Fed is taking the candy away, (c) economic  growth and earnings forecasts are optimistic, and (d) the market is overvalued. 

It has been a while since the last correction: 19% (a near bear) in 2011 and almost 10% in 2012.  We get these on average about once a year, but that doesn’t mean we’re “overdue” as there’s a lot of variability in these time frames.  For example, we had just one correction (11%) during the 1990s.  The Fed is continuing its bond purchase (QE3) reductions with termination set for October, providing the economy is okay.  That is a drag on asset prices, but Chairman Yellen promises low interest rates ad infinitum for an extended period,  a positive for the market.  Regardless, there are arguments for rising interest rates this year and next, so the jury is still out.  Projections for second-half economic growth and earnings could very well be optimistic.  The bond market’s persistent low interest rates are signaling a weak economy, so we’ll see.  We will get a better handle on these as Q2 earnings and forward guidance roll out starting this week. 

Valuation is a controversial subject and whether or not the stock market is overvalued depends on the measure used and how it’s interpreted.  For example, trailing (past year) price to earnings (p/e) is just above the historical average, while forward (coming year) p/e could be somewhat elevated, depending on the “e.”  The CAPE mentioned last week is getting a lot of attention because its current level is consistent with the beginning of severe bear markets in 1929, 2000, and 2008.  As seen in a recent Financial Times article, forward p/e and CAPE are good at predicting market return over the coming ten years, with Shiller currently predicting about a zero overall return.  My own work that’s tested CAPE within the model shows that it does not improve the model’s timing signals.  And the historical data and model tests show that higher valuations are tolerated when interest rates are low.

Moreover, while CAPE says the next ten years will be flat for the S&P, it says nothing about the cyclicality of the index, which is the model’s main concern.  For example, the period between 1966 and 1982 for the Dow and S&P shows a flat secular trend, but surrounded by nine prominent cycles for a timing model to exploit (see chart).  The model, by the way, uses dividend yield (d/p) for the “500” as a much better valuation indicator than p/e.  And keep in mind that fundamentals as a category (which includes valuation) is just one of the model’s four categories of indicators… and the least important at that.  More on the influence of each category over the model’s score in a future post.

So, whether we get or not get a correction this year will at least partly depend on how earnings and the economy might compensate for Fed tightening actions (there’s more to it than these two factors, which is why the model has so many data points).   Meanwhile, the model has its finger on the pulse of  primary trends, keeping in mind that primary downtrends are near  to corrections (in magnitude but not necessarily in time frame).  It sees a small 6% probability that the current  pullback is part of an 8% pullback that would last at least eight weeks.

Jul 6

94.9

Buy

1985

+0.0%

+8.6%

+8.6%

+11.1%

The Dow scales 17k and the “500” bores in on its own 2k, with both indexes climbing to new historic highs at week’s end.  The excuse to go higher this past week appears to have been a positive jobs report and toned down concerns that interest rates will not meaningfully rise in response.  Whatever the offered reasons (an ongoing bull market is reason enough), this bull market remains intact.  Meanwhile, the media celebrates millennial numbers and simultaneously frets over high valuations. 

Millennial numbers such as 17,000 and 2000 have no direct technical meaning regarding indicators, but do have psychological meaning, which indirectly influence sentiment and some technical indicators used by the model.  Scaling new millennials gets investors’ attention, such as “This market is relentless so I must …” (fill in the blank).   Some investors might worry than we’re near the end of the bull and therefore “take money off the table” to preserve some profits (but when do they put it back on the table?); others might expect more highs, thereby renewing a commitment to stocks, regretting not being fully invested or even not invested at all during the current five-year 193% run-up in our benchmark index.  In extreme regret cases we can have a melt-up (see June 22 post).  Regardless, significant stock market reactions are noted by several of the model’s indicators.

And then there’s valuation, typically captured by one of several variations of the price to earnings ratio (p/e).  Of special interest is a high value for Schiller’s cyclically-adjusted p/e (CAPE).  Its currently elevated level is associated with the start of past bear markets of some significance.   I’ll address valuation in an upcoming post, including how the model incorporates this important influence. 

Jun 29

94.1

Buy

1961

+0.0%

+7.2%

+7.2%

+9.1%

Politics is the art of looking for trouble, finding it everywhere, diagnosing it incorrectly, and applying the wrong remedies.

Groucho Marx

Ok, ok, exaggerated to be sure, but entertaining never the less… and, as this is not a site for political commentary, although sometimes unavoidable, let’s substitute “Financial advice” for “Politics.”  Surely the revised quote does not apply to most financial advice, but  true often enough.  (Present company excepted, of course;-)

Happy Independence Day

Enjoy, try not to fret over the stock market on that day.  Really, just one day.

Jun 22

94.3

Buy

1963

+0.0%

+7.2%

+7.2%

+9.2%

Fears ease over Iraq, global oil prices hold steady, Chairman Yellen issues soothing words regarding interest rates, and… the “500” surges to a new historic close.  Go figure: It’s one high after another, with the attendant fears that this high could be the last, the one that marks the beginning of a new bear market.  But then…  (a) ongoing bull markets do make one high after another, (b) it’s still easy monetary policy, so don’t fight the Fed, at least not yet,  (c) there’s enough fear out there that investors as a whole remain underinvested, (d) classic Econ 101 says prices will rise as demand outstrips reduced share supply (thank you Adam Smith), and (e) where else are you going to get return anyway? 

So, this bull continues to hum along its bumpy uphill path, mowing down bears, until it runs out of gas.  And maybe before that happens, just maybe, especially if we first have a correction and absent a seriously negative external event, we could have a market melt-up, an unexpected surge that defies even the optimists, a parabolic rise lasting weeks or months, as a large number of skeptics throw in the towel and BUY, a capitulation by the bears.  Over a five-month time span ending in March 2000, the Nasdaq rocketed 85% (in 5 months!), a melt-up during the dot.com bubble.  The S&P 500 had a mini-melt-up of 22%.  And what followed? A melt-down: 71% for the Nasdaq over the next 18 months; 37% for the S&P.  The Nasdaq has yet to recover from that record close in 2000, more than 14 long years later.  Clearly, exponential increases can become unstable, collapsing as if under their own weight.  Not a good thing, really.  As to the model’s actual (live) performance during that melt-down?  It lost just 10% of its portfolio. 

Now, to be fair, the standard strategy gained just 2% during the melt-up, giving early sell signals prior to the market top.  But… over the nearly two-year period starting at the beginning of the melt-up and ending at the end of the melt-down, buy and hold lost 21% while the standard strategy held the loss to 8%.  Tough times, and more to come with a short 4 month and shallow 21% bull followed by a 34% bear in 2002.  Spanning this entire three-year period (Oct 1999 to Oct 2002 which featured a melt-up, bear melt-down, bull, and bear) buy and hold sustained a loss of 33% while the standard strategy posted a 9% drop.  Note that these are the model’s live results, not optimized back-tested numbers for the current model.

If you’ve had the patience to bear with me thus far, the point is that capital preservation is the hallmark of any good timing system, provided it also capitalizes on most of the substantial gains that follow the hard times, thereby outpacing buy and hold with much less risk.  Moreover, the actual performance of retail investors during this time period was even more dismal, as bear-market fears precipitated selling near bottoms and bull-market regrets encouraged buying near tops.  Or, for many others, fear keeps them mostly out of the stock market, over periods of time so long that the benefits of stock investing are rarely realized.  Not so with the disciplined followers of a good timing strategy.  Dynamic rebalancing is another option for those not willing to follow a timing model, but who otherwise need a disciplined approach that avoids the emotional pitfall of buying high and selling low.

Jun 15

93.0

Buy

1936

+0.0%

+5.7%

+5.7%

+7.0%

Market pulls back about 1% from its 1951 record closing high earlier in the week.  The catalyst?  Spiking oil prices from stunningly rapid territorial advances in Iraq by ISIS, the al-Qaeda offshoot with ambitions for a nation-state that would govern under their interpretation of strict seventh-century Sharia law.  The geopolitical and economic concern is:  How high might oil prices go should there be a major disruption of Iraqi oil supplies, which account for about 10% of OPEC output.  Why worry?  History shows that sustainable oil shocks are followed by recessions.  And it might not take much of an increase this time, given the wobbly global economic recovery from the Great Recession.  The market’s and model’s reactions were muted, so far, so we’ll see what develops this coming week as the ISIS army marches toward Baghdad.  The latest news shows government forces taking back some territory from the militants, a welcome development.  If Baghdad falls, although not likely, the major southern oil fields will be at risk.  Moreover, the formation of a caliphate bent on Jihad with the West is unthinkable.  Think Afghanistan under the Taliban and al-Qaeda, but on steroids, far better organized, equipped, and wealthy, with a fighting force that’s more like an army than a disorganized group of ragtag (although fierce) fighters… and with the will and means to carry out terrorist attacks in western countries.  Or maybe we worry too much?  The next few weeks are key, as we witness military events and mark the reactions of oil prices, stock markets and the model.

Jun 8

93.4

Buy

1949

+0.0%

+6.4%

+6.4%

+8.1%

New record highs were extended this past week on the backs of European Central Bank's easy money moves and a Goldilocks U.S. jobs report. 

The recent sequence of all-time highs makes for good narratives and pads our portfolio while on a buy signal, but as correctly pointed out by a reader, the all-time high adjusted for inflation was back in March, 2000, not the current record close.  For the S&P 500 that 1527 high marked the end of a remarkable 582% 12-year bull market run, followed by devastating twin bear markets that took the index down 51% into 2002.  This past week's high of 1949 is 28% above the peak in 2000.  Since then, however, the Consumer Price Index (CPI) has advanced 40%,  clearly revealing that the current inflation-adjusted high is actually less than the historic high in 2000. Looked at another way, the current high in 2000 dollars is at 1394 v 1527, or 9% below the record close in 2000. And that's using the CPI, which many analysts view as government-biased to show less inflation than what is actually occurring, resulting in less government inflation-linked payouts.  See for example shadowstats.com and its controversies.  Still, highs are highs and we'll take them when on a buy signal.

Yet, the most important question is: What's the value of our portfolio now versus what it was at the peak in 2000?  Are we better off now than back then, even when accounting for inflation? The model's live performance as of this week shows a total gain of 168% since the 2000 peak, well above both the 40% inflation advance and the buy-and-hold 67% total return. These figures include dividend reinvestments, which is why buy and hold outpaces inflation, although not decisively.  Put another way, starting with $100,000 at the record close in 2000, buy and hold ends up with $167,000 as of this writing, versus $268,000 for the model’s standard strategy.  In 2000 dollars these numbers respectively deflate to $119k and $191k.  Clearly, inflation undermines the purchasing power of portfolios over time. The trick is for gains to outpace inflation, to be better off in real terms.  Note that these results based on the model are actual, not theoretical, as described in the Reality Check link… and independently confirmed by TimerTrac (well… since we started getting tracked in 2002).

As stated often in these pages, a good timing model will mostly ride the uptrends and mostly avoid the downtrends, beating buy and hold by a comfortable (not spectacular) margin and, more importantly, with lower risk. Why is lower risk important? It calms down investors' fears, encouraging confidence and the discipline to follow the successful model's buy and sell signals. Study after study in behavioral finance have shown that real investors (versus theoretical buy and holders) all too often end up buying high and selling low, the wicked mirror image of what should be, as emotional investment decisions ruled by fear and greed take their toll on portfolio performances. 

Jun 1

92.7

Buy

1924

+0.0%

+5.0%

+5.0%

+6.0%

Another week another all-time high, this time the Dow joining the “500.”  This bull market is increasingly unloved and mistrusted, with many pundits predicting its demise.  And demise it will, someday.  Probably not yet, though, because the Fed and other global central banks have our backs while economic signals remain mixed, sluggish growth continues, and jobs creation is under par.  Yes, the Fed is easing up on the accelerator, but as long as markets believe it will keep interest rates (indefinitely?) low or re-accelerate if necessary, the bull will likely continue, with a mix of pauses, minor setbacks, and forward movements.  A fly in the ointment? A geopolitical, economic, or other shock that changes investors’ mindsets.  China surely come to mind.  Its global economic clout means that its problems are the world’s problems as well, among these: a financial crash rooted in a housing bubble along with an incredible amount of unoccupied housing inventory; military conflict given its increasing belligerence toward its neighbors and its rapid military buildup and modernization.  For more see the March 30 post and the FAQ on the model’s accommodation to external events.

May 25

93.3

Buy

1901

+0.0%

+3.7%

+3.7%

+4.1%

S&P 500 cracks past 19th century to new all-time high.  Correction talk is out the window, for now.  As insisted by the model this entire year, the primary uptrend remains intact , namely that  we’ve been in a 8%+ up-ramp over at least an eight-week time frame (actually, according to the model, since the buy signal last year following a mistaken five-week sell signal; in reality the primary uptrend began in June, 2012).  The model would have to issue a sell signal if it were to judge the mirror image by anticipating or confirming a 8%+ down-ramp over at least an eight-week time frame.  Note that a “flash” 10%+ correction that ends in less than eight-weeks is not part of the model’s genetics.  This is by design, to ensure a certain degree of stability in the model.  Moreover, most of these rapid corrections are over fast, as the index resumes its primary uptrend.  To be sure, a correction is in our future, we just don’t know when, nor does the model.  (If we did know we would be at the top of Forbes Richest Billionaires.)  The model is not a forecaster in the usual sense; it “simply” estimates at the end of each week the probability (score) that the index, as of this moment, is in a primary uptrend.  If that probability is low enough (39 or below) it decides that the primary uptrend has reversed to the downside.  By definition, a reversal to a primary downtrend would begin at the top of a primary uptrend.

Don’t confuse the map [model] for the territory [reality].

Harvey CORE

At this time the model claims a 93.3% probability (its score) that we remain in a primary uptrend.  Really?  93.3%?  A model is a representation of some reality.  In this case, the model looks at past sequences of up and down primary trends and attempts to reproduce them, first during its development and then live, week by week, as seen here and now.  While the model is mathematically complex, it’s no match for the causal complexity that generates these actual trends.  If not 93.3% then what?  Only the ethers know.  So we judge the model on how effective it is relative to alternatives.  Look at the documented results within this website and compare the year-by-year returns of the model v buy and hold and v your own stock portfolio returns.  Judge for yourself. 

Have you met Harvey elsewhere in these pages? He’s my alter ego, a character in a couple of my books and cases that I would assign to my MBA students.  Click the link and decide for yourself what’s fact and what’s fiction. 

May 18

86.8

Buy

1878

+0.0%

+2.4%

+2.4%

+2.3%

Dow and S&P scale to new all-time highs on Tuesday, then pull back in two-day selloff.  There was lots of media hand-wringing over the selloff, which amounted to less than 2%, with the week ending more or less flat and just 1% below the record that very same week.  The de jour cause for all the angst appears to be the recent and possibly ongoing damage to the momentum stocks from last year’s outsized gains, in particular tech, social media, and small stocks, which have led the overall market in some past corrections.  And, of course, the usual macro fears persist: the hesitant U.S. economic recovery, soft growth in Europe, and what’s going on with the elephant in the room (China, see posting, March 30).  And how will Central Banks respond. 

The Fed can change how things look, but not what they are.

Jim Grant on CNBC

Same old?  It is what it is and will be what will be.  The model listens, assesses, and decides.

May 11

87.2

Buy

1878

+0.0%

+2.4%

+2.4%

+2.3%

Dow ekes out another record high, while the “500” languishes and the Nasdaq continues its selloff.  After last year’s impressive gains, the broad market so far this year is taking a breather, as opposing investors neutralize each other.  The economy does appear to be slowly improving, yet headwinds persist that are fanned by fears over the global economy, geopolitics, the Fed’s continued unwinding of its purchases, and weakening financials, housing, and retail sales.  See Jim Cramer’s take on these issues.  Maybe it’s just a matter of the economy catching up to last year’s outsized gains.  Or maybe we’re heading into another crisis, this time fueled by sovereign debt, the increasing burden carried by municipalities, states, and countries.  See Andrew Ross Sorkin’s explanation based on the Reinhart and Rogoff book This Time is Different. 

Highly indebted governments, banks, or corporations can seem to be merrily rolling along for an extended period, when – confidence collapses, lenders disappear, and a crisis hits.

Reinhart & Rogoff, This Time is Different: Eight Centuries of Financial Folly

Ok, so markets will churn, markets will rise, markets will fall, markets will bubble, and markets will experience crises.  And the more we read, especially the “doom and gloom” folks… the more confused and fearful we get.  I have a sense of objectivity and equanimity by following the model, as imperfect as it is, as are all models, many far too imperfect (present company excepted :-).  It might be instructive to read or reread why I developed the model.

May 4

89.9

Buy

1881

+0.0%

+2.5%

+2.5%

+2.5%

Market advances in quiet trading week, with the Dow scaling a new all-time high.  The S&P is just half a percent below its high last month; the Nasdaq is 18% below its historic blow-off high more than 14 years ago, although this tech-laden index handily outperformed the others coming off the bear-market lows in 2002 and 2008/2009 .  The week included good economic reports regarding manufacturing and jobs growth, although the surprising drop in the headline (U3) unemployment rate was mostly attributed to a drop in the labor participation rate, as discouraged job seekers and retirees left the labor force pool.  The more telling U6 unemployment rate also includes part-time workers and the “marginally” unemployed, those who want to work but who have stopped looking.  This rate is nearly double the commonly reported U3 rate.

Issues troubling the market remain, especially the increased violence in Ukraine, a greater likelihood of military intervention by Russia, and possible military aid to the Ukrainians by the West.  Still, strict sanctions and Russian retaliation through its energy sector are not likely, given the reluctance of Europe generally and Germany in particular not to harm its business interests with Russia.  Doing so could tip Europe into a recession.  Moreover, under the current foreign policy of this Administration, the U.S. will not go it alone.  So, the odds remain that stock markets will not be seriously affected.  Yet, if the situation really gets hot on the ground… and Russian troops overtly cross the border?

It’s that time of year when the media brings up “Sell in May and go away,” also known as the Halloween Indicator.  Google it.  This strategy sells at the beginning of May and usually buys at the beginning of November, or sometimes the beginning of October.  Over the period 1970-2013 the seasonal strategy that buys in November collects about a 10.6% annualized return, including reinvested dividends and “off season” money market (T-Bill) returns.   The strategy that buys in October shows a return of 10.8%.  These compare to a 10.3% return by simply buying and holding.  The half percent difference might not seem like much, but over the course of 43 years we’re talking a 22% greater buy-October Halloween portfolio (11% for the buy-November strategy), thanks to the magic of compounding.  Moreover, volatility (risk) is lower with the seasonal strategy, compared to buying and holding, because the portfolio is out of the stock market five months of the year.  Over this same period the back-tested standard timing strategy gains 17% per year and the aggressive version 23%.  Starting with $10,000, portfolios under each strategy end with $760k for buy and hold, $929k using the October Halloween Indicator, $9.8 million for the standard timing strategy, and $109 million using the aggressive variation.  Back testing showed that incorporating any one of various seasonality indicators into the model degraded the model’s overall performance.  Bottom line?  Let’s follow the model. 

Apr 27

87.0

Buy

1863

+0.0%

+1.5%

+1.5%

+1.1%

Flat week belies intraday volatilities.  Friday’s selloff appears to have been a reaction to disappointing tech earnings (led by Amazon) and increased hostilities in Ukraine.  The G7 nations threatened additional sanctions, without acting, while disagreeing on more serious sanctions that would target Russian economic sectors, rather than just individuals, the inaction likely over concerns of damage to their own economies and wishful thinking hope that Russia will comply with the Geneva accord.  Meanwhile, the Russian economy is weakening further, its stock market is down 17% since February, capital outflow continues, and S&P just downgraded Russian sovereign debt.  Will Putin rethink his military strategy given the evolving economic consequences, especially if sanctions are assertively ratcheted up this coming week?

Last week’s post remains relevant:  Economic reports are mixed and Ukraine simmers, while the government in Kiev is more or less left in the lurch.  I’m not sure that our stock market has priced in a more serious escalation of the conflict, or… maybe it will not amount to much in the end (financially speaking, while overlooking long-term geopolitical negatives and security, international law, and moral hazards)… meaning our stock market (and model) are behaving as if the consequences will be inconsequential… we’ll see. 

Apr 20

86.7

Buy

1865

+0.0%

+1.5%

+1.5%

+1.2%

Market and model rebound, with the “500” posting its biggest weekly gain of the year, up 2.7%.  The index now stands where it did two weeks ago, as does the model.  The pullback last week offered up a small opportunity to beef up lean portfolios… and it will not be the last such window, probably sooner rather than later.  Volatility, both up and down, should continue, with the upcoming week full of market-moving news: many earnings reports, economic data on housing, manufacturing, and durable goods and the Michigan consumer sentiment survey.  And the Ukrainian conflict’s low-boil could spill over, especially given the timid responses to date by the key Western Powers (U.S., G.B., Germany, France), likely further encouraging Putin’s territorial ambitions, as if he needs much encouragement.  An escalation of the conflict will directly impact western (and Russian) economies, assuming the imposition of meaningful sanctions and further assuming military aid to the Ukrainians and possibly other threatened Baltic states.  And stock markets likely will react negatively, at least initially.  Maybe then Russia will back off as its shaky economy implodes.  Nice doesn’t work when opposed by an adversary with a different set of rules and values.  Or the WPs, such as they are, might let Putin have his way, as in Crimea, in which case the markets will yawn and move on to their usual issues.  Bottom line? This geopolitical mess is far from over, including its influence over stock markets.

Follow-up on the High Frequency Trading post on April 6:  Check out this video.  The next day the NY AG’s office pressured BATS into admitting that they also use slower feeds to price trades.  In other words, using fast and slow feeds to price trades favors fast over slow, setting up the likelihood of favoritism on an uneven playing field.  For an interesting and entertaining (as always) analysis by Grant Williams see this.

On a brighter note, completely and refreshingly unrelated to our stock market endeavors, we’ve updated the Travel and Critter Chronicles blogs.  These links are also available within the About page.  Hope you enjoy… and have the time, energy, and means for your own pleasurable adventures. 

Apr 13

68.7

Buy

1816

+0.0%

-1.2%

-1.2%

-2.8%

Market rallies on Wednesday following the release of Fed minutes, indicating it wanted to avoid the perception that it is willing to tighten monetary policy; then came a two-day plunge that brought the S&P to a 4% (ordinary) pullback from its all-time high on April 2.  The Nasdaq’s tech selloff continued as it settled 8% below its recent high.  The previously high-flying biotech sector entered its own bear market at 21%.  The model responds in kind, swooning about 20 points for the week, mostly from a weakening of its technical indicators, in particular relative strength and its version of the death cross.  One of its “euphoria” indicators based on extreme lows had also turned negative several weeks before.

The media is all over this, citing that we haven’t had a correction (10%) in two years and maybe it’s even the start of a bear market (20%).   I don’t expect the latter but would not be too surprised by a quick ten-percenter over the next couple of weeks, especially if we have a blowoff, a couple of days of panic selling that well exceeds those last week.  If so, and the model remains on its buy signal, then it could very well be an opportunity for underinvested portfolios to add stock holdings, especially the tech sector through QQQ and the volatile biotechs using IBB.  I’ve been out of the former and looking to get back in, while remaining in the latter.  And let’s not forget the model’s benchmark index, the S&P 500 and its proxy SPY, my own core investment during buy signals.  We last had a little scare at the turn into February.  See the Jan 26 and Feb 2 postings, which are relevant once again.

Apr 6

88.1

Buy

1865

+0.0%

+1.5%

+1.5%

+1.3%

Fed Chairman Yellen walks back interest rate hawkish comment (see March 23 post) and the S&P 500 scales a new all-time high at 1891 on Wednesday; enthusiasm evaporates by Friday as our benchmark index settles 1.4% under the high, while the Nasdaq is in a steeper pullback at 5.5%, mostly the result of a tech selloff.  Although the “500” ended up for the week, the model eased back, showing some concern over the earlier enthusiasm. 

Have you read Liar’s Poker and The Big Short by Michael Lewis? Terrific and entertaining financial journalism that read like novels, the former on 1980s bond trading, the latter on the 2007-2008 crisis.  The buzzword this week is High Frequency Trading (HFT) as Lewis hits the talk-show circuit to promote his latest, Flash Boys.  So, the game is rigged (what a surprise) as the HFT players game the system to gain milliseconds and pennies by front-running trades that affect everyone else.  Legal?  Maybe yes, maybe no, depending on who executes the trades; the FBI, SEC, and Federal and NY State Attorney Generals are opening investigations.  Unfair?  Sure, but it’s never been a level playing field.  Ethical? Questionable, although we can let philosophers debate an issue made possible by technology.  Does it affect us? Insignificantly, unless a flash crash becomes more than flash.  (See Historical Note, October 20, 2013 post in the archive.) 

I know folks who refuse to invest in the market because it has a dark side… yet, these same folks have given up huge stock market gains by observing and moralizing from the sidelines.  The model looks at the big picture, the domestic macro market as reflected by its benchmark index, the S&P 500, over time periods many orders of magnitude beyond milliseconds.  Sure, skimming pennies (unfair and maybe illegal in HFT), insider trading (illegal), and Ponzi schemes (mega illegal) are examples of market rigging and fraud, but they should  not affect our willingness to participate in just about the only game in town.  Stay above the fray.  Focus on the BIG picture.

Mar 30

94.8

Buy

1858

+0.0%

+1.0%

+1.0%

+0.7%

Market struggles to stay above water year to date and within striking distance of the Mar 9 record high; pullback now stands at about 1%.  Market sensitivity will likely continue to day-by-day events regarding Russian realpolitik and the sanctions (or lack thereof) melodrama, US economic news, Fed policies, concerns over the upcoming earnings season, and a wobbly China staring at a housing and banking bust that would make ours look like child’s play.  An unraveled China would be perilous for world economies.  When it come to pass (all market-based economies eventually boom & bust), let’s hope that it’s in slow enough motion, giving financial systems (and the model) time to assess and adjust.  For a good read on this topic see China’s Minsky Moment.

…over periods of prolonged prosperity, the economy transits from financial relations that make for a stable system to financial relations that make for an unstable system.

Hyman Minsky

Mar 23

94.9

Buy

1867

+0.0%

+1.5%

+1.5%

+1.4%

Market rebounds, encouraged by some good economic news and little concern over surprisingly-weak (so far) sanctions imposed by the US, Europe, and Russia over the Crimean land grab.  Meaningful sanctions going forward could have consequential reactions in the stock market, as noted in last week’s post.

The market did have a hiccup on Wednesday as the Dow swooned 200 points over Fed Chairman Yellen’s unintended transparent comment at her first press conference that she could see interest rates rising some six months after ending the taper sometime in the fourth quarter of this year.  The market’s knee-jerk reaction to an earlier than expected increase was typical, but far from warranted.  First, rate rises will likely be slow and generally indicate an improving economy.  Second, it’s good for savings such as money markets, bank accounts, and CDs, thereby providing more capital.  Third, the model has its finger on this particular pulse.  The influence of rising interest rates depends on other factors as well.  At what point are we in the interest rate cycle?  Where are we in the business or profit cycle?  Is market momentum strong?  Is the market overvalued, undervalued, or fairly valued?  How much is emotion influencing the market at this time?  Reality is much more complex and subtle than the exercise of considering just one or two indicators, prominent as these may be.  The model includes 7 interest rate data points (out of 26 overall) and generates a holistic score using 16 indicators (monetary, technical, sentiment, and fundamental) that are calculated from these data points.  Yes, interest rates that rise excessively do negatively impact markets, but the extent of this effect is either moderated or enhanced by the status of other relevant indicators.

Mar 16

93.8

Buy

1841

+0.0%

+0.0%

+0.0%

-0.7%

Market pulls back 2% from last week’s “500” historic high.  Investors’ concerns are not going away for now and more volatility this year should be expected (see last week’s post).  Actually, worries are part of the game, which is why bull markets climb the proverbial “wall of worry.” 

The worry du jour is Russia’s occupation of Ukraine’s Crimea (all but annexed with today’s vote), Putin’s intentions on Ukraine’s eastern territories, and the extent of Western financial sanctions and Russian counter moves.  And what eastern European country previously under Soviet control might be next? Why does the market worry?  “It’s the economy, stupid,” to quote James Carville’s famous line during Clinton’s first presidential campaign.  Ukraine is small potatoes, economically speaking, a basket case in fact.  But… natural gas pipelines crisscross its territory from Russia into Europe, which is about 30% dependent on its rogue supplier.  Cutoffs or reductions have big implications for a society’s financial well being.  Once more we see the value of energy independence through conservation, renewables, and, yes, fossil fuels...  and the importance of imports from friendly and stable suppliers for additional energy that’s needed.  Besides the energy issue, western banks and corporations are significantly invested in Russia.  The cauldron’s ingredients of sanctions, energy, banking, debt, equity, and currency exchanges make a  toxic brew.  And that affects interconnected global financial and stock markets.   But then again, if sanctions are mild then markets will likely be little affected, or even relieved, assuming the absence of military confrontation.

Geopolitics clearly do affect markets, so it’s worth paying attention, especially how the model assesses market reactions to external events.  For now, the model (through its diagnostics) says the patient (market ) is strong, and can take some blows (downside volatility).   For now.  As long as the model remains on a solid buy signal, pullbacks are usually opportunities to commit more funds to stocks, for underinvested portfolios willing to deal with turbulence (risk).  Should the model rapidly weaken, while remaining on a buy signal, it might be prudent to stay put and hold off on new investments.

Mar 9

97.5

Buy

1878

+0.0%

+2.0%

+2.0%

+2.4%

Market swoons over Ukraine on Monday, picks itself up, and ends week with another S&P 500 record high at 1878.  The payoff since March 2009 for riding this bull? Ignoring dividends, 178% for the “500” and 153% for the Dow.  The winner by far is the Nasdaq at 289%.  The Dow sits less than 1% from its recent all-time high, but the Nasdaq is still 14% below its magical 2000 high, 14 years later.  Imagine a Nasdaq buy and holder over that time span, if there is such a character.  Has yet to breakeven from that debacle. 

Those betting against riding the bull say the ride is getting long (maybe, see last week’s post), price to earnings is too high (it’s been much higher), especially the Shiller variation (my work shows that it’s better for secular than cyclical bulls), profit margins are shrinking (but the economy could still expand),  it’s losing Fed support (not by much, they still have our back), the cold war is back (it never left, just hibernated).  But… following net withdrawals from stocks since the 2007-2009 twin bears went into a long sleep, investors poured money into the market last year, the highest net inflows on record.  And let’s not forget that the Super Bowl indicator tells us this will be a good stock market year.  Ok, ignore that last comment.  For now, the bull appears to have energy.  The negatives cited will likely cause more downside volatilities of the 5 to 15% variety, so let’s hang on and let the model spectate.

Mar 2

96.9

Buy

1859

+0.0%

+1.0%

+1.0%

+0.8%

Bull market reconfirmed as the “500” posts a historic high at 1859, despite a late-day selloff on Friday in reaction to the now ongoing Ukrainian crisis.  At 60 months and 175% this cyclical bull stands tied for 4th in length and 5th in return, the latter just short of fourth place.  Does it have further to go?  Sure, it could, historically speaking.  Is it about over?  Sure, it could be, historically speaking.  We will know when the time comes.  Meanwhile, I’m counting that the model has our backs as it seeks to identify the beginning of the next primary downtrend, the next inflection point that marks a change in the primary trend, a likely condition that would mark the beginning of the next bear market.  The model is about 97% “certain” that we remain in a primary uptrend, thereby expecting a continuation of this bull.  See also last week’s posting.

Now for a segment on Mojena’s Believe it or not.  The TimerTracked medallion bottom left links to a professional service that confirms and ranks the actual performances of several hundred timing services and their multiple strategies.  This link shows the solid outperformance of the model v the S&P 500 from 2002 (when they started tracking us) through 2013.  The standard strategy returned about 50 percentage points better than the benchmark, while the aggressive strategy upped the result by 90 points.  Entertain yourself by playing with the parameters on that page.  Over the difficult five-year period starting December 2005, during which the S&P 500 lost 2%, our standard strategy returned a confirmed 74% based on the Nasdaq 100, the 7th top performer out of more than 500 that were tracked, and with just six trades, far less than the six timers that were ahead of us.  In another citation, we’re ranked number one based on bookmark frequency for market timing.  Who cynically said that “you get what you pay for with a free service?”  This site remains free of charge and ads.

Feb 23

92.7

Buy

1836

+0.0%

-0.3%

-0.3%

-1.0%

Market treads water as we remain in a pullback, 0.7% below the 1848 year-end high and 5.4% above the pullback 1742 low on Feb 3.  The model estimates a 93% probability that we remain in a primary uptrend, within the current bull market.  A bull market that’s up 171% from the 677 low in March, 2009.  A bull market that at five years is a bit long in the tooth, falling at about the 75th percentile for length and return (see table in FAQ).  A bull market missed by far too many individual investors, who were devastated and frightened by the dot-com bust bear markets in 2000-2002 and the financial-crash bear markets in 2007-2009.  Buy and hold anyone?  Fortunately, the model avoided the worst parts of these bear markets and, just as importantly, got back in to reap most of the subsequent bull market gains.  True, the model missed buying at the bottoms and selling at the tops, but was close enough to significantly benefit over theoretical buy-and-holders.  I say “theoretical” because I don’t believe very many true buy-and-holders exist in the real world.  Serious bear markets (the 50% variety) have a way of flushing them out.  For me, the model gave me the confidence (courage?) to get back in following very scary times.  Without confidence, frightened investors simply stay out and miss the big run-ups.  Until, that is, they feel comfortable enough to test the waters, all-too-often late within the existing bull market.  Is that happening now?  A recent WSJ article cites some evidence that this is the case, especially among day- trading small investors, who show a significant rise in both trading activity and margin debt.  If these activities get out of hand or should we have a market melt-up, then the current bull market might be near its end.  A necessary sign for this to happen would be the start of a primary downtrend, an event that’s monitored and, in theory, detected by the model at its inception. 

Feb 16

91.4

Buy

1839

+0.0%

-0.3%

-0.3%

-0.8%

Rally off pullback low gains steam as index settles near its all-time high; model adds to last week’s surge.  More to go?  Market does not appear overbought short term.  A proprietary indicator within the model that measures standardized deviation from a moving trend (for you quants out there, an exponentially smoothed standard deviate based on a moving mean and moving standard deviation) shows that levels above its current value are common, occurring about 23% of the time.  Bollinger Bands are another way to look at this.  Note how the ETF proxy for the “500” (SPY) in the chart tends to (but not always) bounce off the top (overbought) line.  The lower (oversold) line suggests buying opportunities. 

Risk-off trading appears to be back and, let’s face it, the stock market might be the only game in town, as it was last year.  The Fed still has our backs.  And interest rates on savings and bonds are practically non-existent (“thanks” to the Fed), thus discouraging traditional investment alternatives to stocks.  Still, we do remain in a pullback, until the 1848 high is broken.  Interestingly, gold (GLD) is rising once again, while the  beaten-down gold miners (see GDX for example) have soared recently. 

Feb 9

85.9

Buy

1797

+0.0%

-2.6%

-2.6%

-4.2%

Monday’s swoon is more than erased by the week’s close; model likes the action,  leaping 25 points.  The pullback reached nearly 6% on February 3 and now stands at about  3% from the year-end all-time high of 1848.  The model’s technical indicators firmed up and its sentiment reading improved as pessimism rose during the steeper part of the pullback.  It’s been over two years since we had a correction, so we could say that possibility is still in play given that we remain under the high.  The model, however, sees this as less likely than last week’s reading, now assessing that probability at about 14%.  I like the 6:1 odds that we remain in a primary uptrend.  As usual, the operating strategy for underinvested portfolios is to buy pullbacks, especially those between 5% and 10% and while the model remains on a buy signal, assuming tolerance for greater volatility in the portfolio.

Feb 2

61.1

Buy

1783

+0.0%

-3.4%

-3.4%

-5.3%

Volatile week ends lower.  We remain in a pullback, reaching 4% midweek.  On average, since 1970, we get 5% to 10% pullbacks two to three times a year and corrections breaching 10% about once a year.  Since the bear market low in 2009 we’ve had 8 pullbacks (one just shy of 10% in 2012) and 2 corrections (16% in 2010 and 19% in 2011), both severe, the latter a deep one that almost tipped us into a bear market.  Meaning that over the past four years or so we’re right on schedule with pullbacks and behind the curve on corrections.  Will upcoming years then overcompensate corrections, giving us a reversion back to the mean?  Who knows.  It’s a question of timing and we have the model to help us on that issue as the market unfolds.  At this point the model’s judgment is that the current pullback is not consistent with a correction.  Nothing is certain; it’s all about uncertainty.  Given the model’s current score, we have about a 39% chance (in theory) that we are in fact within an evolving 8% pullback of at least 8 weeks based on Friday closings (a primary downtrend).  I’ll go with the more favorable odds that we remain in a primary uptrend.

Jan 26

60.5

Buy

1790

+0.0%

-3.0%

-3.0%

-4.7%

Two-day knockdown takes the “500” 3% below its year-end all-time high of 1848; model responds in more than kind, plunging 30 points.  Troubling, yes, but a one-week drop and pullback of this magnitude are common.  Moreover, a one-week or two-day decline does not make a downtrend.  While revising downward its probability that we remain in a primary uptrend, the model nevertheless maintains its buy signal.  Far more often than not in the past, pullbacks less than 10% while the  model remains on a buy signal present opportunities to buy lower for portfolios that are underinvested.  Still, a steep pullback near 10% or correction beyond that is possible and maybe even overdue.  The model now bears closer scrutiny, as it has fallen into a sensitivity range (21 to 82) that makes it vulnerable to rapid changes. 

Jan 19

89.6

Buy

1839

+0.0%

-0.4%

-0.4%

-0.8%

Flat market for the past month shows resilience.  A one-day knockdown is followed by recovery.  The model is on solid ground, meaning that the intermediate uptrend (weeks to months) looks good.  I’m more concerned about the secular trend (years to decades), given US and global high debts and deficits, the inevitability of significant increases in interest rates, and degradation of economic freedom.  A recent article and survey show the US dropping out of the top ten to 12th on an index of economic freedom.  Hong Kong, Singapore, and Australia lead the pack; Canada is 6th, Ireland 9th, the UK 14th, Germany 18th, Japan 25th, Spain 49th, Mexico 55th, France 70th, Italy 86th, Brazil 114th, China 137th, Russia 140th, Venezuela 175th, and Cuba 177th, just ahead of North Korea at the back of the pack.  See also the April 28 posting in the 2013 archive and the second FAQ regarding cyclical and secular markets. 

Regular readers are aware that the model’s concern is far from secular, as its main focus is the detection of a new primary trend of plus or minus 8% over two or more months.  Still, it does react to secular factors in the sense that the amplitude and frequency of a primary trend and cyclical bear market are influenced by the secular trend.  For example, a primary downtrend and cyclical bear are more likely during a secular bear market.

Write-ups and downloadable files for the 2014 model are ready.  Click links in the menus at left and top.

Jan 12

87.3

Buy

1842

+0.0%

-0.2%

-0.2%

-0.5%

The model for 2014 is now operational.  Compared to last year’s, the new model trades a bit less and generates higher scores;  it would not have given the 5-week switchback last August, costing the standard timing portfolio a 4.6 percentage point deficit for the year v buy and hold.  On average the new model trades about 4 times over 3 years, versus the old model’s 3 signals in 2 years.  New sell/buy bands are 39 & 76 v the previous 38/73.  In short, the revised model is more stable, yet improves overall performance since 1970.  Look for details in new write-ups over the next couple of weeks.

Jan 5

63.4

Buy

18310

+0.0%

-0.9%

-0.9%

-1.4%

Year ends with new all-time highs for Dow and S&P, the latter at 1848; the first two days in the new year give up nearly 1%.  The given model score is based on the 2013 model.  I’m now working on the revised model for 2014.  It should be operational within the next couple of weeks.

 

 

 

 

 

 

 

 

 

 

ARCHIVE

2013

2012

2011

2010

 

 

NOTE

The TimerTrac link at left is a free report provided by an independent company that tracks the performance of market timers. Note that the report does not account for dividends and their reinvestment, as we do, and as would be the case for reported returns in the media, thus showing lower returns for both buy-and-hold and the standard strategy during buy signals than those seen under the live performance table in our Reality Check page.  The difference over long time horizons can be significant, as reinvested dividends make up about 30% to 50% of total S&P 500 returns, depending on the chosen time period.  Also note that the timing model is a statistical mathematical model that issues buy and sell signals.  The strategies (standard & aggressive) are the trades that are made when these signals are issued. 

Distribution
Copyright © 2014 Richard Mojena. All rights reserved. All materials contained on this site are protected by United States copyright law and may not be reproduced, distributed, transmitted, displayed, published or broadcast without the prior written permission of Richard Mojena at mojena.com. You may not alter or remove any graphics, copyright or other notice from copies of the content.  You may download or print one machine readable copy and one print copy per page from this site for your personal, noncommercial use only.

 

Disclaimer

Specific and personalized investment advice is not intended by this communication. Its contents are for the public record as a free public service. Information is based on the analysis of past data and assessments by the models. Future performance may not reflect past performance. Profitable trades are not guaranteed. No system or methodology ensures stock market profits. No guarantee is made regarding the reliability or accuracy of data. In other words, use this stuff at your own risk!

Front Page      Timing Model      Reality Check      FAQ      Downloads      About      Mea culpa?

mojena.com

As is, no spin