PM End of Week Market Commentary - 1/31/2014

davefairtex
By davefairtex on Sat, Feb 1, 2014 - 5:51am

Gold finished Friday up +1.10 to 1244.40 on moderate volume, silver was up +0.06 to 19.18 also on moderate volume.  The gold/silver ratio fell -0.15 to 64.86.  GDX was down -0.04% on moderate volume, and GDXJ was off -0.17% on moderately heavy volume.  Gold actually rallied in Asia and London up to 1255, but right before the NY open the rally stopped, reversed, and gold ended up closing mostly flat.

For the week, gold was down -24.40 [-1.92%], silver off -0.72 [-3.59%], GDX -0.80%, and GDXJ down -2.40%.  The senior miners outperformed everything else in the PM complex, turning what would normally be a distinctly bearish tone into a more complicated picture - part bullish, part bearish.  This is visible in the charts as well - gold has corrected, while GDX is tracking sideways with an upward bias.

US Equities/SPX

Currently, it seems that the equity market is tied in with PM.  Over the past few weeks, rallies in the equity market have tended to lead to sell-offs in PM, and vice versa.  At least for right now: equities good = gold bad.

So since we have this correlation, perhaps we should look at the equity market for clues as to where gold might be going.

According to a trader I respect, "distribution days" are one good indicator of an upcoming market correction. 

The "distribution day" signal is all about spotting high-volume selling, which likely comes from big funds, bank prop desks, and institutions.  Once these guys bail out en masse, markets tend to drop.  So to spot distribution days you must examine volume, and see if the volume on the down days is significantly greater than the volume on the up days.  In some sense, its about tracking money flow.  Is money flowing into, or out of the market?

Looking at the chart of SPX below, we can see 6 distribution days over the past few weeks.  One or two distribution days are normal in an uptrend over a 4 week timeframe, but a cluster of 6 distribution days within a few weeks is a definite danger sign.

You can also see the 20 EMA about to cross the 50, which is also a bad sign, and the 50 MA itself looks ready to move down as well.

So if you read Hussman you know the equity market is overvalued, overbought, and overbullish, but that sort of condition can remain in place for quite a while before anything untoward occurs.  But we also now have two sequential bites at the Tapering Apple.  Last time tapering happened, the market had a nasty hiccup.

And when this situation is then paired with a technical trader's bearish assessment of money flows, well, trading is all about "going with the odds" and at this point, the odds appear to be favoring a move down for equities in the near future.

Which according to our current understanding is bullish for gold.

Miners

Senior miners mostly tracked sideways this week, while juniors corrected modestly.  The chart picture for the mining shares still looks bullish - price of the miners leading a rising 20 EMA gradually up towards a 200 MA crossing, with the 50 MA flattening out rather than continuing to drop.

In addition, down-day volume was relatively lighter than the up-day volume, which says there is no  distribution occurring.

Buying individual mining shares can be risky, however.  One of the larger miners, Newmont Mining (NEM) reported record gold production for 2013 and promptly sold off -10.37% on massive volume, making a new low for the year.  Investors didn't like NEM's forward guidance for 2014, which was lower than expected.  NEM yields 3.7% at its current price.

The USD

The USD rallied steadily this week, with the bulk of the move coming the day after the FOMC meeting.  The buck was up +0.82 [+1.01%] closing the week at 81.37.  It is nearing the 81.50 resistance level; a break above this level would likely lead to more buying of the dollar, and it would be bearish for gold.

If problems persist in the emerging markets, we should expect a stronger dollar over time.  But its also possible that emerging market issues will also result in safe haven flows into gold, so the picture is perhaps more complicated than simply "dollar up/gold down."

The moves in the buck are all about capital flows.  When capital flows in, the buck rises.  And when the capital flows in, it must "land" somewhere - bank deposits, bonds or equities (or possibly real estate, etc).  Right now, equities are dropping, and bonds are rising.  Since the buck is rising too, that suggests money is flowing in from overseas and landing in the bond market.

Rates & Commodities

The 10 year treasury rates continued their fall, and are down to 2.66%.  20 year treasury rates down to 3.35% - down from a high of 3.72% at the start of 2014.  This amazing rally in the US treasury bond market no doubt makes the Fed happy, although the move is likely not due to tapering but rather currency flows from the rest of the world towards the US, as well as a move out of equities and into bonds.

Commodities overall were mostly flat, tracking sideways but with an upward bias.  As we know, rising commodity prices tend to be good for PM.

However, copper had another bad week.  It has driven through 3 moving averages without stopping, and appears to be heading for 3.10 at a minimum.  The price drops are accelerating as time passes.   "Something bad" happened to copper about two weeks ago, there was no rebound at all, and things aren't getting better, they're getting worse.  From all I know, copper = China, and the copper market is telling us that there are problems in the Middle Kingdom that likely haven't been corrected by the last-minute rescue of "Credit Equals Gold #1 Fund" http://www.businessinsider.com/china-shadow-banks-2014-1 by some mysterious third party.  (The irony of the name: "credit equals gold" - could it be more perfect?  No, credit does not equal gold, and it is possible copper is our canary in the coal mine, just now tipping over in his cage.)

Physical Supply Indicators

* Shanghai Gold Exchange is closed this week for the Chinese New Year celebration.

* The GLD ETF gained +2.70 tons of gold this week, and is back up to 793 tons.  After dropping pretty consistently since the April gold crash, GLD ETF has neither gained nor lost gold over the past month.

* Registered gold at COMEX rose 2.16 tons, to 13.68 tons of gold - still low, but improving modestly.  First notice day for February gold contracts was January 30; February is typically a relatively large delivery month, and 13 tons is not a particularly large amount of gold.  It will be interesting to see what happens.

* ETF Premium/Discount to NAV; gold closing (15:59 close price) of 1244.70 and silver 19.155:

    PHYS 10.38 0.00% to NAV [up]
    PSLV 7.66 +2.53% to NAV [down]
    CEF 13.39 -5.70% to NAV [down]
    GTU 43.65 -5.38% to NAV [down]

Discounts on the ETFs mostly increased, but by relatively small amounts.  PHYS was the exception where premiums increased modestly, moving PHYS back to flat.

With Shanghai closed, we can only judge by COMEX and GLD, which suggest gold is not in short supply.  The physical ETFs agree with this picture overall.

Futures Positioning

The COT report is as of January 28th.  Managed Money increased net long exposure by a big 17k contracts - mostly by short covering.  Producers decreased their net long exposure by a similar amount, mostly by selling off their longs.  Looking at the price charts, most of this change likely happened last week; the producers (the big banks?) cashed in on the move above 1265, while managed money covered their short positions.  This ties in with my theory of why the banks like those big spikes up: they can use the big volume produced by Managed Money short-covering when resistance levels are breached to sell their supply of COMEX futures they accumulated way back when.

Producers are no longer net long but still look quite bullish, while Managed Money short exposure is still  high - 62k contracts that should provide fuel for the short-covering rallies should gold continue moving higher.

Moving Average Trends [20 EMA, 50 MA, 200 MA]

Gold: short term DOWN, medium term DOWN, long term DOWN

Silver: short term DOWN, medium term DOWN, long term DOWN

Gold's correction on the day after the FOMC meeting dragged the 20 EMA back into a bearish downtrend.  Gold still remains above its 50 day MA, however, which I believe is more important.  Still, PM is now in a downtrend in all three timeframes once again.

Summary

This week mining shares looked strong, when you compare them with the performance of the metal.  Senior miners held steady, juniors dropped a bit, while gold and especially silver sold off.  Momentum has shifted to the downside, especially in silver, which looks dangerously close to a serious breakdown and shows no sign of rallying.

Looking at the various ratios and averages, gold and silver are both in a moving-average downtrend in  all three timeframes.  GDXJ:GDX is tracking sideways (neutral), while the picture remains clearly bullish with GDX:$GOLD.  The gold/silver ratio hit a new peak this week above 65 (bearish) which underscores just how weak silver really is - we not far from the July 2013 high in the GSR of 67.

After ending the 5 month downtrend last week, gold has backed off, supported by its 50 day MA but seemingly not yet ready for a move through 1280, which is the new cycle high.

With Shanghai closed, COMEX registered rose, and GLD's tonnage has risen as well.   ETF buyers were modestly negative this week.  Physical demand is mildly negative.

Gold's uptrend remains in place but it has corrected somewhat, silver is looking ready to break down through 19, and the mining shares tracking sideways provides the tiebreaker that says - more likely than not we move higher from here regardless of tapering.  The US equity market entering a distribution phase confirms this for me.  It may not happen next week - gold may have to drop further before the COMEX buyers appear - but I think the next move up in gold (with silver possibly lagging behind) is not so far away, and it should coincide with the US equity market making a move south.

 

7 Comments

KennethPollinger's picture
KennethPollinger
Status: Platinum Member (Offline)
Joined: Sep 22 2010
Posts: 654
Goldman, the name of the GAME?

FYI

Life is really weird sometimes.
 
Check out Floyd Norris's article yesterday in NYTimes, Business,
:Dark Side of Capital in Emerging Markets, " especially last two paragraphs--just what Martenson and many others are predicting:
 
"It took forever and then it took A NIGHT."
 
Also two more: 1) "Goldman Deal Threatens DANISH Government"
and 
2)"Libya Says Goldman Didn't Explain Options"  sounds like their work in Greece.
 
Is Goldman headed for the peak of the pyramid???  The lead of the
New World Order??
 
Scary shit
cmartenson's picture
cmartenson
Status: Diamond Member (Offline)
Joined: Jun 7 2007
Posts: 5752
Excellent Commentary Dave

Dave,

Your weekly summaries are just excellent.  So thank you for them!

Here's my question: Given today's market weakness (S&P 500 is down 25 as I type, to 1757, breaking the 1770 support I've been tracking pretty convincingly) do you think the distribution you noted is over, and, if so, what are some logical targets for this move down?

davefairtex's picture
davefairtex
Status: Diamond Member (Offline)
Joined: Sep 3 2008
Posts: 5463
thanks chris

Here's my question, given today's market weakness (S&P 500 is down 25 as I type to 1757 breaking the 1770 support I've been tracking pretty convincingly) do you think the distribution you noted is over and, if so, what are some logical targets for this move down?

Well, I can't answer this in just a word or two.  (Have I ever been known for brevity?  Perhaps I need an editor...)

Hmmmmm.  No, distribution continues.  If you look at price action intraday, every rally is being sold.  And the high daily down-volume pattern remains in place.  I'm just guessing, but I suspect today will be a high volume down day.  It looks like traders just want out right now.

As far as price targets, I'm not that clever.  That's Louise Yamada, she's the one to ask about those sorts of things.

My technique is to try to see how the market behaves when it gets to particular support levels.  As you pointed out, it seemed to slice right through 1770 - triggered apparently by a poor ISM report at 10:00 EST.  Did I say poor?  I meant awful.  Here's what econoday has to say:

http://mam.econoday.com/byshoweventfull.asp?fid=461307&cust=mam&year=2014&lid=0&prev=/byweek.asp#top

Today's ISM report is signaling very significant slowing in composite growth for January, at 51.3 for a sharp 5.2 point decline from December. This is the lowest reading since May 2013 and the sharpest monthly drop since May 2011.

The bad news is centered, unfortunately, in new orders which are down a very steep 13.2 points to 51.2. This is one of the largest monthly declines on record. If there is solace, it's that the plus-50 rate of 51.3 rate still points to monthly growth, just at a much much slower pace than December.

Ok, so the market is sucking pretty badly right now, and it was apparently triggered by this news item.  So reading the tea leaves, this is what I'm seeing right now - more negative than positive.

Negatives:

1) Down days continue to be higher volume than up days.  That's a bad sign.  Institutions are bailing out, and any rallies are being sold.  As long as this is happening, we (likely) continue going down.

2) We have established a pattern of a lower high and a lower low.  That's the sign of a downtrend.

3) The market couldn't even rally back to the 50 MA.  All it could do was consolidate at the 1770-1800 level.  That just seems quite weak to me.  A market that can't rally will likely drop further.

4) Market is now clearly news driven; no more "any news = market goes up."  This is a big change from just a month ago.  Again, its never the news, it's the market's reaction to it that matters.

5) In the past, tapering has led to a market drop.  Two bites at the tapering apple by the Fed (whoa, Mom & Dad really mean it this time!) implies psychology likely has traders looking downhill right now.

Positives:

1) Small caps are hanging on better than the large caps.  QQQ hasn't broken down yet.  That's still risk on.

2) The selling isn't (currently) disorderly.  Meaning, it doesn't appear to be a panic.  It can get a whole lot uglier than this.  Some of us may remember how things were back in 2008.

3) The uptrend has been in place for a long, long, long time.  Uptrends don't die easily.  You must respect the trend in place.  Until I see that 50 MA cross the 200, I'm not writing off this uptrend, and neither will the longer-term focused traders.

A great trader I respect once said, "when things get confusing, pull back to the weekly charts to get a better sense of what's going on."

Looking at the weekly chart, I see that the SPX has found support on its 50 week MA the past two dips.  Right now that's at 1688.

So first clear point to watch is 1688 - that is the 50 week MA.  Next point might be 1550, which was a breakout point that was tested and held on the way up, and the next support level which looks to be quite strong (it is also the 200 week MA) is 1400.

Here is what I'm going to watch for:

1) watch the market's reaction to the different bits of news that come out.  If the market rallies on good news, that's fine.  If it sells off on bad news, that's ok.  If it starts to sell off (or rallies only weakly) on GOOD news, that's quite dangerous, and likely signals significantly lower lows ahead.  Gold did that last year.  Boy was that outcome ugly.

2) See if the market can rally meaningfully.  If we break the pattern of lower highs, the "buy the dip" folks will come out of the woodwork.

3) See if the up-day volume (or a flat-day volume) starts to be larger than the down-day volume.  That means accumulation is happening instead of distribution.  That's the institutions loading up, which is usually bullish.

From my trading perspective, I'm just watching.  I was hoping to short the rally to the 50 MA, but the market was so weak, I didn't get my rally...

One thing to remember too.  Markets typically don't drop off a cliff in one day.  They (generally speaking) move down in stages.  So if we rally tomorrow, the important things to watch are:

1) what is the volume of the rally days vs the volume of the down days?

2) can we exceed today's high?

HughK's picture
HughK
Status: Platinum Member (Offline)
Joined: Mar 6 2012
Posts: 761
Six sigma error by mainstream economists
davefairtex wrote:

  As you pointed out, it seemed to slice right through 1770 - triggered apparently by a poor ISM report at 10:00 EST.  Did I say poor?  I meant awful.  Here's what econoday has to say:

http://mam.econoday.com/byshoweventfull.asp?fid=461307&cust=mam&year=2014&lid=0&prev=/byweek.asp#top

Today's ISM report is signaling very significant slowing in composite growth for January, at 51.3 for a sharp 5.2 point decline from December. This is the lowest reading since May 2013 and the sharpest monthly drop since May 2011.

The bad news is centered, unfortunately, in new orders which are down a very steep 13.2 points to 51.2. This is one of the largest monthly declines on record. If there is solace, it's that the plus-50 rate of 51.3 rate still points to monthly growth, just at a much much slower pace than December.

Ok, so the market is sucking pretty badly right now, and it was apparently triggered by this news item.  

Dave and all,

Regarding the surprisingly low number for the Institute for Supply Management's factory report, there is a nice article from ZeroHedge that points out how the actual ISM number was six standard deviations below the average number forecast by economists.  

ISM Has Biggest Miss on Record, New Orders Plunge Most Since 1980

I trust many others here at PP have already seen that article.  What impressed me most about it was the giant gap between what the economists who were polled expected and the actual data.  This six sigma chasm between the forecast and actual data seems like it would support the hypothesis that conventional macroeconomic paradigms are becoming outdated.

While many here may respond by saying that's very obvious by now that mainstream economics has some major theoretical dissonance, I don't think it's at all obvious to most people.  

I am currently re-reading Popper on what makes a theory scientific.  He says that a scientific theory must be falsifiable.  That puts many, if not most, social science theories into the category of what Popper calls pseudoscience, although Popper acknowledges that "sociological or psychological" theories can meet the criterion of falsifiability.  I will try to post more about this later, and in a more appropriate thread, but I'm interested in how we here shape and reshape our social science theories.  To what extent are we in the process of falsifying mainstream economic theories that no longer seem to do a good job describing our emerging reality and to what extent are we holding our own theories up to falsification?

I would posit that Dave has a beneficial influence on our thinking by using technical analysis to contain and qualify - if not necessarily falsify - beliefs/opinions/theories that many of us share about market behavior in the precious metals complex.  Note: I'm not trying to start a big debate with that statement...I'm just interested in applying Popper's falsifiablity criterion to our social science (mostly economic) theories here at PP and among the limits-to-growth crowd as a whole.

In any case, I still find the six sigma gap between the economists' predicted number for the ISM factory metric and the actual number to be one data point that does not support conventional macroeconomic paradigms.

I also continue to be very grateful for your commentaries, Dave.  Thanks.

Hugh

Hrunner's picture
Hrunner
Status: Gold Member (Offline)
Joined: Dec 28 2010
Posts: 256
The Bigger Picture

HughK,

Interesting post about pseudoscience, and some responses from an actual scientist.

Let's frame the discussion with some basics.  True scientific methods use a centuries old process of observe, formulate a hypothesis about a law or how a scientific / natural / social (eek) process works, building on prior science that is foundational and "solid", also forged from some reasoning and logic, then devise an experiment to "prove" or "disprove" a hypothesis.

That "disprove" part is what I think you are referring to as "falsification", which to me is just another way of saying it.

Usually, important scientific ideas, or hypotheses, can be distilled down to deceptively simple statements.  "Since gravity bends space, a massive object with lots of gravity will appear to bend a light beam".  "DNA, as double helix, is replicated by using one strand as a template to form the new strand".

These are clear, precise statements.  Part of good science is crafting a hypothesis with clarity and precision, not leaving a lot of wiggle room, which leads to meaningless and ambiguous conclusions- not helpful.   

So science is kind of a confrontational sport- perhaps that is why I appear confrontational to Mr. Dave at times, I am seeking as much clarity as possible, because good theories don't allow for 15 possible outcomes or explanations.  A good scientist will devise a good experiment that is essentially binary- the data carefully controlled, carefully collected, and many times replicated data confirm or disprove your hypothesis.

In truth, the vast majority of hypotheses get modified around the margins, but the underlying truth is intact.

I really struggle with social science, though am not dismissive, because it is so hard to design experiments that are controlled, repeatable and valid. 

I don't even consider economics a science.  I think it is best characterized as a descriptive discipline that seeks to create a framework to view the 'money world', but that's about the best I can do.  That's why it is laughable that anyone, the Bernankes and Krugmans of the world, think they can craft glorious "models" that accurately predict the future.  True hard laws of science decribing heat transfer, voltage and current, the behavior of molecules in homogenious solutions, can rely on accurate on 'models' or formulas. 

"Economics" cannot.  For a couple of simple reasons.  Economists are trying to create a predictive model that, to be accurate and thus useful, must account for in economy hypotheses, among other things, millions of currently unpredictable things like the weather (don't get me started about predicting climate), natural forces such as volcanoes, earthquakes, and many other phenomena which, even though they may conceivably be one day predictable, are generations away from any level of measurability and predictability.  And often the interaction of billions of variables with billions of other variables.  And if that's not bad enough, economist must create models that predict human behavior.

Sorry, can't be done.  That's the fundamental nature of free will.  It is a voluntary choice.  We don't know why people chose to marry one person over another.  Why you like red and I like blue.  Why you like jazz and I prefer hip-hop.  And millions upon millions of free will choices that billions of individuals make each day.  From very consequential people like heads of central banks to smaller fry like myself. 

I repeat, these variables are unknowable, if not completely stochastic.  Buzzzzzzz.  Game over.  Economists and their models are trying to model the unmodelable.

These are critically important points to understand because the very foundation of our society grows out of some important world views of humanity and economics.

If you want my historical context on economics, and how it is relevant to us today, I see the whole central planning themes of communism, socialism, fascism (like we have in the U.S. with the collusion of big corporate entities with the government) as growing out of two sources. 

One, a political world view that the majority of humanity is uneducated, unwashed, stupid and downright scary and dangerous and thus must be controlled and "helped" or terrible things will happen.

If people are allowed to have guns, they will run around like crazed animals shooting each other until everyone is dead.

If people are allowed to price money without government control, awful things will happen like business cycles, poverty, and job loss.

And two, the false notion that economics is just another branch of science.  The belief that the "science" of economics can replicate the fantastic advances that real science such as electromagnetism, chemistry, physics and biology have given us.

Therefore, since economics is just another science, we need to come up with our own datasets, theories, and models so we can join the Age of Science.

A key part of this world view is that, despite all of these animal subhumans running around shooting each other and pricing money willy-nilly, we can achieve a utopian society, if only we have enough control by enough smart, elite people in power.  With enough glorious economic models.

We should note that the collectivist-central planning view is in direct opposition to the American Founders view that the best and highest society is a fundamentally moral society, underpinned by access to proper education, Biblical principles and values, and stable family lives.  This free society does not need to be controlled in any way by a massive, oppressive government armed with economic models.  And given a small government that serves a small, but necessary utilitarian role, a measure of prosperity and generally good living will follow, with the individuals and freely associated communities making their own choices about how they want to spend their time and money.

The best economies are the freest economies.  Minimal regulation, with government involvement limited to few things like (possibly) collecting and publishing data, serving as a third party to retain documents and make them available, and perhaps a handful of rules about monopolies etc.  We are clearly way past that point.

Free economies set correct prices for things- by definition, the correct price is what a buyer is willing to give a seller.  Free economies work because they (not always in ways that may seem 'scientific') value things that are scarce and devalue things that are common.  Including energy, precious metals, great paintings, and Stradivarius violins.  Free economies and markets bring out the best in human creativity and work because it is rewarded appropriately.  Free markets may do crazy things (crazy to you, not to others perhaps), and are not "perfectly efficient", they are simply the best game in town.

Free markets do not need models, theories, and most especially, economists and central planners, to work.

H

 

davefairtex's picture
davefairtex
Status: Diamond Member (Offline)
Joined: Sep 3 2008
Posts: 5463
extremistan, evidence, and TA

HughK-

I also continue to be very grateful for your commentaries, Dave.  Thanks.

Thanks Hugh.

Regarding the surprisingly low number for the Institute for Supply Management's factory report, there is a nice article from ZeroHedge that points out how the actual ISM number was six standard deviations below the average number forecast by economists.

I don't fault the economists.  Herding behavior is why the market is Extremistan, and why Taleb said statistics as a means of analysis of human trend behavior was just doomed to failure.  We have far too many six sigma events - so is the fault in the people, or in applying statistics and its inherent assumption of a normal distribution to Extremistan?  To my mind, statistics is just the wrong tool to use.

A rule of thumb of traders is: "the public is right about the trends, but wrong at both ends."

Again Taleb would probably talk about Fat Tony - who most likely uses rules of thumb - and Dr John who tries to apply statistics.

My approach is to look for clues, and I don't pretend its scientific.  That's why I call it "reading the tea leaves" - so neither I nor any reader gets any ideas as to repeatable science actually being involved.

I do try and keep things evidence-based as possible, and for me, evidence means price & volume data, ratio charts to analyze relationships between markets and sectors, and moving averages.  This is primarily because I'm subject to the usual bouts of wishful thinking, seeing what I want to see, smoking hopium, confirmation biases, etc, that everyone suffers from and these tools impose a regularity that forces me to see the price movements the way they actually are, rather than how I'd like them to be.  But I think the interpretation of the bits of evidence that are fed into the rules of thumb - and the selection of the rules themselves - is ultimately art, not science.

If I were to take a truly scientific/model-based approach, here's what I'd do.

First I'd get historical customer trading data for a good cross-section of the trades that go across my bank's data feeds.  I'd link these trades to my bank's customer profiles, and then I'd try and come up with a model that tracked what each group of customers did based on various events, given what I know about psychology of markets and the like.  I'd likely come up with some sort of computer learning algorithm that, over time, could predict what certain groups of people would do given specific price movements.  It is possible that with enough data, enough training, and a good enough model, I could actually predict (using representative realtime customer trading data as input) what the market would actually do given specific price events.

Lacking the model, the training, and the realtime data - I am stuck with Fat Tony's rule of thumb and art.

Lastly - I know what you are saying about "its all manipulated" vs my price/volume TA approach.  A trader friend once told me, "if the markets are manipulated, the manipulations will show up in price & volume, and all those rules of TA will still apply."

If the conductor of a train is trying to deliberately run you down, or the conductor is blissfully unaware of your presence, does it matter?  Regardless of which story is true, when the train comes, it is probably best to move out of the way.  And if TA shows you when the train is coming - I'll happily ignore all the theorizing about what that conductor's evil plans are and focus on what the train is doing instead.

In the long run, it all blows up because of the energy issue, and in the shorter term, its the debt issue.  But between now and then - I believe watching the train to be a more useful approach rather than speculating on the thoughts and motivations of the conductor.

 

HughK's picture
HughK
Status: Platinum Member (Offline)
Joined: Mar 6 2012
Posts: 761
Thanks, Dave

Dave,

Thanks for your reply.  Just briefly, before I hit the hay...

I don't fault the economists.  Herding behavior is why the market is Extremistan, and why Taleb said statistics as a means of analysis of human trend behavior was just doomed to failure.  We have far too many six sigma events - so is the fault in the people, or in applying statistics and its inherent assumption of a normal distribution to Extremistan?  To my mind, statistics is just the wrong tool to use.

I will chew on this idea.  On one hand I understand that the normal distribution might be the wrong model for markets.  Jim Rickards prefers the power law, for example, with it's much greater tail risk.  But, that's also a statistical model.  It seems that statistics can be a helpful tool for describing or even trying to predict behavior of large groups of people, but maybe that's wrong.  I haven't read any Taleb, but now I want to, as all of this applied directly to my central question, which is basically, "what are we doing in the social sciences?"  Are all of our theories just art, as you describe TA?  Are all they nothing more than narratives which may be more or less appealing, but none of which have any deeper validity?  Of course, I don't think that's true, but I think it might be partially true.  Popper suggests that social science theories can be scientific if they are falsifiable, but there may be non-falsifiable approaches to social science that are also valid.

Anyway, thanks for responding; very wholesome food for thought.

Cheers,

Hugh

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Login or Register to post comments