I listened to Bernanke talk about this once. The academic view seems that things weren't that awful until 1931, when a large Austrian bank failed (Google Creditanstalt), which triggered a wave of global bank failures [1]. The resulting lack of credit was the cause that made the depression Great. When Creditanstalt went down, Dow was off ~50% from its 1929 high, which still happens every now and then. We went through a similar decline during the 2007-08 financial crisis and came out all right thanks to aggressive and globally organized central bank activity. Unfortunately such an organized response didn't happen in the 1930s so the economies kept contracting, and the Dow eventually ended up losing 90%. Check out the charts in that period, it is fascinating.
I've heard other somewhat contrary things, one is the birthrate started falling well before the '29 crash indicating that working class and especially farming communities were actually not doing well economically. You can look up graphs if you want, but the birthrate drops relentlessly from 3.2 in 1920, to 2.2 by 1930. As the economy improves post 1935 the rate picks up again.
My take on recessions and debt is to consider that when an entity (family, business, etc) pays off debt they are foregoing some consumption. Key thing. Now in normal times foregoing of consumption frees up economic output that is invested in the real economy. As I pay off my mortgage, the bank reinvests that money in building more houses. A virtuous cycle, depending how much you like concrete, sprawl and freeways.
In a debt driven recession there is a problem, the economy isn't limited by available production. There is a lack of demand. Because as people frantically pay of debt they are foregoing consumption that doesn't free up more physical or labor resources for investment. It just creates a lack of demand. The money instead goes to correct balance sheets which is something that exists on paper (or in modern times le computer machine).
That's the situation many parts of the developed world are in today. The solution is for the government to step in an manually rebalance consumer and commercial balance sheets. A political problem exists when the people in charge can't let go of the 'paper wealth'. And that's the current issue.
It seems the indices are down 30% or so from their prior highs, which is unpleasant but nothing like the 1929-32 period. Even Nasdaq had cratered far worse in the early 2000s. By comparison, Dow Jones went from 381 to 41 in just about 3 years from its 1929 high, which is an astonishing 90% loss. Actually I didn't know it was that bad until I looked it up.
As of the open today, ChiNext was down 41% in a month.
As of right now, 90% of China's entire stock market is frozen. It's absolutely a 1929 style disaster. Were the authorities not stepping in constantly, the carnage would be far worse already - which is another way of saying, eventually it will get there no matter what they do.
The 10% limits on falls each day are horrible this way. It actually creates a continually falling market, we won't know how bad this crash really is for awhile.
The Chinese stock market hassufefred rampant manipulation to ramp it up, and is now correcting itself. Even with the steep drops so far it's still well up over the last 12 months, after all. All this means is that those rises were not justified and are being wound down. The government's panicked response isn't helping, and the fall may well be more severe than the economic realities justify, but what's new?
There is no severe immediate risk of a depression in China. Their stock market is relatively very small compared to their economy as a whole and Chinese firms as a whole are nowhere near as dependent on it for raising capital as in western countries. Therefore the ammount of damage any fall in the stock market can do to the Chinese economy is fairly low.
The Chinese leadership has staked a lot of it's credibility and prestige on the rises in the stock market, as evidence of it's good economic management. As a result they, and anyone wantign to curry favour with them, were hyping up the market and blowing air into the bubble as hard as they could. Bad mistake. They'll learn.
I hope you're right. But reading about firms that are raising loans with their stock as collateral. .. shall we say while staying true to your point are perhaps nevertheless even more exposed than your model's worst case. Only time will tell.
Sure, you're right to be concerned, it's no picnic. Some companies will go bust, and some investors will lose their shirts. It will hurt. It's not yet a systemic risk though. Of course the more the Chinese government panicks and tries to manipulate the market back up again, the worse it will get but right now it doesn't have to be all that bad.
Also interesting to contemplate the influence of WW2 in that fact. I think I generally hear that WW2 was a big booster of economic growth, so how long would it have taken without the war?
A glib way of answering this is to look at how much of U.S. GDP went to military spending. It was extremely high during the war and has remained at a higher average level ever since - thus Eisenhower's "military-industrial complex" remarks at the end of his term.
Minus that spending, it's not clear what would have happened, since we never had a return to the old status quo. The military investments of the war motivated the early development of Silicon Valley - thus, we'd be looking at a present day where "somewhere else" might have gained the tech hub crown.
That said, the current moment is one where the premises could change again. China has a strong and growing influence in the world, just as the U.S. did coming into the Depression. The particular circumstances differ, but a broad "cycle of empire" is one way to interpret things.
I think it's a great example of why it's important to try to not judge history. The complications in this one scenario are just so fascinating, yet it's important to resist trying to tie it up in a neat bow.
I use kdb/Q at work and it’s a fun tool to play with so long as someone else is paying for it. It is quite common in finance (and comparably uncommon outside of it). It’s very expensive of course, and the learning curve is hard. In fact, there are plenty of businesses that have sprung up around kdb that offer consultancy services to help you get started. In an unusual maneuver, one of these consulting businesses actually ended up buying majority of Kx Systems, the developer kdb/Q. Anyway, if you know your Q and C++, you will always have a job in finance.
Part of the reason why it’s hard to learn (unless your job depends on it so you are forced to persist) is that the syntax is very terse. Check out this Java API for example: http://kx.com/q/c/kx/c.java. Yes, that’s the actual code you copy-paste into your Eclipse to get started.
I read the whole letter and loved the tone of it! It jumps at you right away that: (1) Here's a guy who really cares about his business, its future, his employees. (2) He talks candidly about the reasons for closing up: his declining health, sales volumes, even including awkward sentences like "So chalk is more environmentally friendly, I think." I am now a fan of Hagoromo.
It was rather refreshing at a time when I can't stand reading more than a paragraph of a typical press release of a BigCo written by lawyers or PR specialists. Unlike this essay, those are intended to obfuscate, not to inform.
Yes. There's none of this maximize-shareholder-value, church-of-the-mba religion that pervades Western business culture right now.
He wanted to serve people sustainably. He did it for decades, and when that became untenable, he responsibly searched for a way to minimize the waste and harm caused by him exiting the business.
I especially love that he treats his machines almost as people, people intrinsically valuable to him. Whereas here, we often treat our people as machines, machines that are eternally disposable.
Not awkward, the sentence you call out. Beautiful, rather, and whether it is an artifact of translation or not, no matter. You didn't mean awkward in a pejorative sense; this is just an affirmation.
Of course! What makes this letter effective and relatable are all those little things that wouldn't have passed the ordinary editorial, legal, or PR filters, if it were written by a typical corporate CEO. If I was teaching at business school, I'd have my students read this as a model of effective stakeholder communications. I think Mr. Watanabe is a shokunin (you know, in the Jiro Ono sense) of the chalk business.
Yes, there’s a "frequency spectrum" of sorts and HFT is at the short end of it. Here, your models monitor and react to trading tick-by-tick, and execution speed is paramount. It’s also where you have the fiercest arms race for the fastest systems, colocation, FPGAs, etc. Next up comes what’s often called “statistical arbitrage,” where you have models that no longer look at tick-by-tick trading but may look at what happens at 30-second, 1-minute, or 5-minute windows. Here you have more interesting relations emerge between stocks and the market, e.g., what does IBM do relative to the tech index, or relative to MSFT, etc. (Such cross sectional relationships don’t seem to matter as much in HFT.) Actually stat-arb was the domain where the earliest statistical approaches such as "pairs trading" emerged. Next up come models that trade daily (or less frequently), and here you begin to see the long-short market-neutral relative-value type of approaches, where some quantitative [mutual] funds may operate. Next up will be the traditional mutual funds, and beyond that you have your Warren Buffett’s, etc.
One thing to keep in mind is that the higher your trading frequency, the smaller the price moves you can hope to capture, which limits how much capital you can deploy in your models. This is why HFT models are usually small in size but have high Sharpe ratios. As you reduce your trading frequency, you can expect to capture larger price movements and deploy more capital but you’ll also be exposed to more of the vicissitudes of the general market, so your Sharpe ratio will decline. Market participants usually carve themselves a happy spot on this frequency spectrum and stay there. I don’t know of any firm who is successful at every spot.
I think you are misconstruing holding period or predictive horizon with latency sensitivity. Many HFTs are looking at statistical relationships like the ones you mention to compute a fair price for making markets. The only trades where HFTs hold positions sub-second on average are pure arbitrages. Like you mention, there simply isn't enough price movement within that timeframe to generate a profit.
All the inputs to their pricing change rapidly, so their order prices must change quickly as well, but they can end up carrying risk for long periods of time. The Australian regulator looked at HFT activity in their markets, mind you probably less sophisticated than US stocks, and found the average holding period was 42 minutes: http://tabbforum.com/opinions/hft-concerns-are-overstated
I'm not sure the average holding period is a useful thing to measure. For eXample, if you are trading a spread it's how quickly you put on the second leg that matters, and who cares how long you hold the pair for. So, 42 minutes has nothing to do with it.
I work in quantitative finance and find that basic probability and statistics help a lot in this field. In fact, they are almost like prerequisites for hiring.
That said, I think everyone would benefit by reading a bit about these subjects, whether or not they’re developers. We seem to have a built-in tendency to underestimate the effect of randomness in life (see Fooled by Randomness and How to Lie with Statistics). To counteract that, I find statistical methods (e.g., hypothesis testing, confidence intervals, etc.) to be very useful tools to keep in mind. Having basic familiarity with such tools may not necessarily make us better programmers but it'll probably make us more rational decision makers in the general sense.
This degradation in comforts of air travel is very regrettable. Especially when it comes to seat pitch, coach class is clearly denied what is necessary and, worse still, this is intentional. There's a lovely quote by a French engineer named Jules Dupuit going back to 1849, which explains the conundrum. I think he was talking about railroad cars but the argument applies to air travel just the same. Quote:
"It is not because of the few thousand francs which would have to be spent to put a roof over the third-class carriage or to upholster the third-class seats that some company or other has open carriages with wooden benches ... What the company is trying to do is prevent the passengers who can pay the second-class fare from traveling third class; it hits the poor, not because it wants to hurt them, but to frighten the rich ... And it is again for the same reason that the companies, having proved almost cruel to the third-class passengers and mean to the second-class ones, become lavish in dealing with first-class customers. Having refused the poor what is necessary, they give the rich what is superfluous."
Snapchat is doing well. I sometimes get on the bus around the same time when the local high school is dismissed, and watch in amazement all the kids pull out their phones and work through their accumulated snaps for the day. Some have hundreds. It's a 45-minute ride to my stop and they’re still watching when I get off.
Few months back I got jealous and installed Snapschat myself. It conveniently scanned my contacts, and found that none of my friends are on it. (I’m late 30s.) To this day, the only snap I received is the default welcome message they send to everyone.
They’re doing a good job of confining their appeal to their target age group. The minute I and my peers appear on Snapchat, it's time to get worried.
And then there's snapchat.com, which is another enigma. I consider myself a reasonably competent technologist but Snapchat makes me feel like an ape trying to figure out a mysterious monolith.
This is why I think their valuation is less absurd than it seems. Their rate of adoption among college age and younger demographics is astounding. Way faster than facebook, or twitter, or any other social network was adopted.
I think a lot of people are skeptical because the concept seems really dumb. But I've just kind of accepted that social networks get popular for very subtle reasons, some of which are non-technical.
Building a social network seems to be more of an art than a science.
Most people are sceptical for the simple reason that the social networks that shed most of their user base long before ever threatening to make a reasonable profit actually had far more lock-in. If Snapchat's user base - a demographic not exactly noted for their long attention spans - gets bored or sick of ads once Snapchat actually manage to sell them in significant numbers, they're not leaving anything behind when they download the next flavour of the month.
The terms attached to this probably make their chance of selling out to Facebook for ~$20bn slim too.
Early Facebook had the facility to get in touch (or stalk!) people I had no contact details for provided I knew their name, and check out upcoming party details, not mention a photographic record of my university life complete with comments and status boosting likes. Facebook didn't need to remain cool to be relevant and useful, and I logged in far more often than someone sent me an actual message from the service - still do a decade on.
Snapchat is a messaging app which distinguishes itself from the other messaging apps users also have installed on their phones by the fact it doesn't preserve any past interactions. Much like when I stopped bothering to log into MSN Messenger because people sent me messages in other ways, it dies in the eyes of its users once the daily updates stop, or even more quickly if the daily updates become near-exclusively advertising of the unwanted kind
Perhaps, although snapchat is deliberately oriented towards a lack of history. I could move my messaging off facebook easily enough, but photo albums of vacations less so.
Smartphone and technology adoption is higher than it was in the 2000s. Most new social networks will be adopted faster than Facebook was in the early 2000s. I believe even Pinterest at a point was "growing faster than Facebook."
You're sure they aren't watching one of the featured stories about a city or an event? At a 10 second maximum, that gives a minimum of 270 snaps - not too impossible but I think there might be more going on there if you're not too familiar with the platform.
Yes, that's another possibility, other users' stories. But I would equate that to reading my entire Facebook news feed every day - I don't even think that's possible anymore with the current mysterious sorting algorithm.
A Hippocratic-like oath by itself would be useless without a powerful professional organization watching out for our interests. If I take the moral high road and get fired by my MBA boss, I have no course of appeal. If an MBA were to fire a doctor for refusing to compromise on ethics, I'm sure the AMA would unleash hell on the MBA. Same for lawyers and ABA. Actually those two organizations make it difficult for MBAs to manage their members so they have two safeguards we lack. (I've seen this in the bank I work. The lawyers report to other lawyers all the way up to the general counsel, who reports to the CEO and the board. Such structure makes it easy to keep ethical conduct high priority without fear of retribution from the MBAs.)
In Ontario Canada we have an organization doing so, the PEO. Any professional that has the word "engineer" in his title, including software engineers must do a license exam, old a valid engineering degree and have 4 years of supervised work experience with references. It isn't as common for software guys then let's say civil engineers.
But what happens in the above scenario? Does the PEO kick ass when necessary for its members? Does it, like the lawyers and doctors, ensure that its members are remunerated well?
It doesn't adjust the remuneration of it's members (nor do lawyers in Canada), it really is "market" based, as for doctors the government has a bigger say since our healthcare is universal.
The PEO will take action against non-members using the title engineer, or take action against engineers not acting lawfully.
One can, and one should, but we as a society have a vested interest in providing incentives to make doing the right thing easier than doing the wrong one.
I've criticized the abrogation of responsibility elsewhere in this thread, I'm not giving anybody a pass or an excuse, but the reality of the cowardice of normal people necessitates that we provide some measure of cover for the people unwilling to risk their necks.
I agree with you both. "One can, and one should," and it’s even better if the profession provides a mechanism to remove from practice its own worst offenders. The AMA/ABA examples I gave above do not just offer protection to their members, but they also hold power to send their unethical practitioners to the poorhouse. (If you’re a junior physician with lots of student loans but lose your license due to an ethics violation, I don’t know what you’ll do. Loans don’t go away in a bankruptcy.) So those professions come with a terrible downside for flagrant violators. We don’t have any such downside in software, which may explain some of the moral depravity mentioned.
I'd say that lack of references is more often a sign of ignorance than innovation. I see that every day; transformative work like John Nash's thesis only come along once in a while.
The way we publish research is broken. So much prior art goes unpublished because it is never accepted by a journal. Negative results, meh results, and flawed research all have value. Proper citation is often difficult but can be as important as the work itself, other wise the knowledge graph is fractured.
You're not in academia, are you? There're hundreds of obscure journals and conferences with very low standards. As long as you choose venues that are within your league, you can publish anything that remotely resembles research, and it will show up on Google Scholar. A new academic paper is published every 15 seconds, and most of them are "negative results, meh results, and flawed research".
Most of them are negative results, though usually not overtly presented as such. In my experience, it's extremely rare for a researcher to throw away months of hard work because the result is null. They almost always find a way to augment it or spin it, write up the paper with the phrase "more research is needed" in the conclusion, and submit to a mediocre venue.
[1] https://fraser.stlouisfed.org/docs/meltzer/bermac95.pdf