This sort of scare tactic means nothing without an effort to explain the correlation. Without a testable theory, it's meaningless data mining.
My guess is someone on the inside has already shorted the market just before press time, anticipating the effect the article will have among unsophisticated investors. And guess what, boys and girls? That kind of "insider trading" is legal -- you can say anything you want in the press, and you can position yourself to benefit from your own article in advance. It's all perfectly legal.
Shorting the S&P500 in the hopes that a MarketWatch article will have a significant effect on the price despite the $14T market cap and high liquidity would be an absurdly risky and low-return way to make money. That's why people who run these sorts of scams focus on penny stocks with low market caps, relatively unsophisticated investors, and low liquidity.
Not that I disagree that the article is meaningless data mining--I just think stupidity is a better explanation than conspiracy.
Edit: I should have used the DJIA instead of the S&P500, since that's what's referenced in the article, but the same points hold.
I think that if this chart does have predictive power, the implications are far stronger and weirder than just "the stock market is going to drop next month". If there is a match to these random ups and downs, then why? What does each of them signify?
i HATE "news" like this. we know it's not correlated, but it's scary how the media can essentially manipulate the market. hope this doesn't get covered by more popular news outlets.
It would be because of the pattern in this chart IFF it spooked enough people. Unlikely.
It could be because of the same things as caused it in 1929 - the pattern in the chart is (extremely weak) evidence of similarities, and so we probably should up our estimates of that likelihood by some epsilon - but (without rigorous analysis) the appropriate epsilon is probably small enough that it's below the precision available to human reasoning. Ignore it or investigate further, don't do anything based on this.
I would want to see what happens to r2 when you drag the 1929 data across the last 80 years of the DJIA or whatever other index you want before I get excited about a 0.9 result.
If you look hard enough in a large enough data-set you will eventually find a correlation. It is just like the experiment where a dead fish placed in an MRI machine was asked to identify the emotional state of a person in a picture. Guess what happened.
The pattern is fairly well established at this point. I am not sure exactly what it means and I'm not sure the obvious interpretations that most people would have are correct, but it definitely argues for some sort of pulsing macro business cycle.
My own hypothesis is that gold's value vs. paper/stocks is a decent indicator of "fear" -- people flock to commodities like gold, silver, and real estate, as well as to bonds, etc. when they don't believe the market is sound. Thus gold's value rises during these times relative to paper currency, which depresses the value of the stock market when priced in gold. During times of hope/exuberance, the opposite occurs. People flee static investments for dynamic ones. Who wants to own a lump of metal when the markets are hot?
There hasn't been a really big macroeconomic "growth story" since 1999, thus this graph.
If the pattern continues, this graph argues the same thing. Given that this is a repeating and established pattern, it's a much stronger argument than the correlation in the original article up top. It argues that we have not yet "hit bottom" in the current macro cycle and that one more crash of some sort lies in the near future before the economy resets itself for the next growth cycle.
> but it definitely argues for some sort of pulsing macro business cycle.
Not really. Without an explanation, it's three bumps having no significance. If you watched a sequence of coin toss outcomes, some heads, some tails, and called the heads outcomes "market surges", you would see the same kind of pattern, but without any real meaning.
Remember that to a scientist, the default assumption is not that an observed pattern has significance, but that it doesn't -- it's called the "null hypothesis".
There's a world of difference between describing a pattern, and explaining a pattern.
Don't repeating patterns increase the likelihood that the pattern is non-random with each repetition?
If you saw a graph of coin flips that showed a clear sinusoidal pattern and that pattern continued over many cycles, wouldn't it be reasonable to at least suspect that the coin was not "fair"? One cycle, maybe not. Two, okay, I'm slightly suspicious. Three, four, five?
Of course the pattern says nothing about causation, just that something is biasing the output somehow.
I am not a statistician but I do have a fair amount of background in things like information theory, and I find it very hard to believe that regularity has zero significance. Plug that output into Shannon's equations and compare it with a random source. Or doing VBR mp3 compression on /dev/urandom vs. a piece of music.
The fooled by randomness argument can be taken too far. In a reductio ad absurdium you could argue that I do not exist because theoretically a random source could produce what I just typed. You would theoretically be correct. If you output digits of pi long enough, they will contain this text.
> Don't repeating patterns increase the likelihood that the pattern is non-random with each repetition?
1. No, not for finite-length patterns anyway.
2. Your question is deeper than it appears, and it alludes to some complex theories about randomness.
Let's say I present a series of decimal digits that is a million digits long, and I claim that they're random, in this context meaning they have no exploitable internal order (i.e. have high entropy). If you understand the risks in making assumptions about randomness, and in spite of all sorts of apparent non-random sequences within the list, you might say, "the list isn't long enough to draw any conclusion about randomness." And you would be right.
My million-digit list might actually be a sequence of digits of Pi starting at some arbitrary point within Pi and extending for a million digits from that point. The list appears random, but it isn't -- it actually has very low entropy.
On that basis, guess how many digits you would need to be assured that they represent a random sequence? Spoiler: an infinite number.
All the principles that apply to a claim of randomness, apply to a claim of non-randomness, and for the same reasons.
> I am not a statistician but I do have a fair amount of background in things like information theory, and I find it very hard to believe that regularity has zero significance.
Okay. If I flip a fair, unbiased coin and record all the flips, counting heads as 1 and tails as zero, over time I will see any number of apparently significant patterns, and the longer the sequence, the more likely that I will see "significance".
In a long sequence of flips of a fair coin, for any particular sequence of length n to have >= 50% probability requires 2^n flips. For example, this means after 256 flips, the chance for an uninterrupted sequence of 8 heads has risen above 50%. Therein lies the problem -- with a large enough data set, you will see all sorts of meaningless patterns.
The only way to sort this kind of thing out is to:
1. Adopt the "null hypothesis" -- meaning start out assuming that a correlation means nothing, then gather evidence to contradict that assumption. In other words, don't start out by assuming what you must prove.
2. Consider alternative explanations for the tested outcome -- avoid confirmation bias.
3. Remember lex parsimoniae, otherwise known as "Occam's razor", the precept that the simplest explanation tends to be correct. And the simplest explanation is that a pattern that lacks an explanation, also lacks significance.
In the final analysis, one cannot make any reliable statement about a data pattern without knowing how the sequence was generated, in other words the result data ultimately has no meaning, only knowledge of the generating function can offer that kind of significance.
> If you output digits of pi long enough, they will contain this text.
Yes, but that's an argument against the meaning of perceived patterns, not in favor of it.
This gets really philosophically deep if you keep going... :)
Like... how do you bootstrap epistemology? Given what you say above, how is it that a completely naive learner learns anything? If you immerse a naive learning entity in random noise, it will only learn random correlations. But if you immerse it in an environment with structure, we must assume it would begin to mirror that structure in its internal state. (Learning is information transfer.) But at some point it has to start somewhere... to start by attempting to correlate one apparently non-random pattern with another.
BTW, I do agree that the graph I posted doesn't prove anything. But if it continues to repeat, at what point should we start questioning the null hypothesis and searching for underlying causal factors? Does statistics have anything to say about that?
This gets into areas like "if we built an autonomous space probe, how would we program it to look for 'interesting' things? Define interesting..."
> Given what you say above, how is it that a completely naive learner learns anything?
Well, that's a very good question, and I think the answer is by being naive, meaning suspending disbelief until the person has enough experience to be an informed consumer of ideas.
> If you immerse a naive learning entity in random noise, it will only learn random correlations.
That's true, but children are natural scientists, naturally curious, predisposed to think there's a mechanism behind everything. If that instinct succeeds, they will look for and sometimes find actual mechanisms -- where they exist.
> But if you immerse it in an environment with structure, we must assume it would begin to mirror that structure in its internal state.
That's true even when the structure is an illusion, as with religion and fixed belief systems. To me personally, the hardest part of growing up is not discovering the real mechanisms of life, but unlearning the phony mechanisms that we tend to be force-fed as children.
> But at some point it has to start somewhere... to start by attempting to correlate one apparently non-random pattern with another.
I would have said that the start is locating a plausible mechanism for a pattern that might otherwise mean nothing, then proving a correlation. Then offering the explanation to one's friends to see if they can find a flaw in your reasoning. Hmm -- I just described about 80% of modern science. :)
> BTW, I do agree that the graph I posted doesn't prove anything. But if it continues to repeat, at what point should we start questioning the null hypothesis and searching for underlying causal factors? Does statistics have anything to say about that?
Yes, it does -- it's the same with all apparently nonrandom sequences. Unless the observer tries to find and test explanations, the default assumption must be that, no matter how persuasive, the data are random and lacking a cause-effect relationship.
Here's my favorite example of what can go wrong. Let's say I'm a doctor and I think I've cured the common cold. My cure is to shake a dried gourd over the patient until he gets better. The cure might take several days but it always works. It's repeatable. It's falsifiable (it might fail, but so far it hasn't). Other laboratories successfully replicate the experiment. So it's "scientific", at least according to the definition of science that doesn't require things to be explained (as with psychology).
To a mature, skeptical mind, everything is wrong with it -- no attempt to explain, confirmation bias, etc. But to someone starting out in life, to someone not sufficiently skeptical, it's a scientific breakthrough. It's not random. :)
Someone told me that part of the reason for the great returns of the stock market in 2013 was because of the fed's quantitative easing program. They said that the easing was increasing the money supply, and that money has no where else to go except into the stock market, because all other investments haven't had great returns.
Is there any merit to that? I was planning to keep investing in stocks because of that. But, really I know nothing.
By doing so, you'll beat the post-fee, ex ante performance of actively managed mutual funds[0]. You're doing exactly the right thing--investing in broad-based indices rather than trying to beat the market. As your time horizons get shorter, you can move some money to bonds (treasuries, bond ETFs, etc.) to decrease your risk (and reward).
Bear Stearns failed mid-March 2008. Despite this, on March 27, 2008 Mark Hulbert gave serious weight to Richard Band's prediction of an "uptrend that could carry the blue chip indexes to all-time highs by late 2008 or early 2009. Dow 16,000 here we come!" http://www.marketwatch.com/story/dow-heading-for-16000-richa...
Following this advice, you would have bought near the top of the market, when what you really want to do is buy low and sell high...
Hulbert is probably the most knowledgeable student of DOW history. He knows that what happens to the DOW in the short run is mostly meaningless noise. He should take his own advice:
"Some psychological researchers, for example, have presented series of random numbers to human subjects and then asked them whether there are any patterns in the data. Overwhelmingly, the subjects claim to have detected patterns in the data.
"We should keep this research in mind as we watch the stock market day in and day out. We may think we have detected a pattern, but that doesn't mean that it really exists."
The y-axis tells the story. In 1928-29, the market lost about 45% of its value (375 down to 200). If the pattern holds, the current market will go from about 16400 to 12400, which is 25% of its value. That's still bad, but not as bad as 1928-1929.
I'm pretty skeptical of charts like this. On the y axis, the scales aren't relative. The gain on the black line before the drop is about 187% (200 - 375). The gain on the red line is 132% (12400 - 16400). On the x axis, there isn't even a label for red so it impossible to tell how the different time periods compare.
> Folks smarter than me: is it prudent to move my assets into cash for a couple of months?
No, bad idea. Don't make moves based on press articles like this -- they're either created by people who want to exploit your reaction, or they're followed by contrarians with no connection to the article, who do the opposite of what the article suggests, and thereby also exploit your reaction.
I've been a doomsayer ever since the 2008 crisis (crises) exposed certain systemic flaws in finance and investing, and I've also been wrong while everyone else fed at the QE trough and did just fine.
While I don't believe that QE regimes and cooked statistics will necessarily end well, I also don't believe this chart proves anything. If you do, however, look into ultra-shorts like SDS and DXD, and if you happen to be correct, you can profit handsomely from such an elevator-shaft drop.
1) it doesn't (alone) tell you what sort
2) it takes quite a number of examples to be very certain of anything
3) you have to be weighting counter-examples correctly, too
4) you shouldn't forget your priors
My guess is someone on the inside has already shorted the market just before press time, anticipating the effect the article will have among unsophisticated investors. And guess what, boys and girls? That kind of "insider trading" is legal -- you can say anything you want in the press, and you can position yourself to benefit from your own article in advance. It's all perfectly legal.