Interesting bits of finance? Well, of course it's a bit subjective. There's a lot of things that look like they're technically interesting, such as various complex derivatives, but I think the real value in those is having a sales guy reeling in the client. In fact, a HUGE number of bank/finance jobs seem to have this characteristic.
The real interesting stuff is devising your own trading strategies. You're trying to find something that orthodoxy for many years thought impossible. And most likely, anyone who found good contrary evidence didn't publish. So it's pretty thin what you have to start with; a field where your own discoveries are actually your own, even though probably other people have come up with similar ideas (and aren't telling). Having no orthodoxy is great for seeing whether you actually understand how statistics work, and in the wider picture, how one discovers anything.
By the way, with your background, there's a good chance you'll be shoehorned into some derivs type thing. You'll definitely be able to do it, although there's some new math. It's initially interesting (pick up the autobiography of Emmanuel Derman), and definitely pays well, but ultimately loses it's lustre. I think the gold rush from science to Wall Street has had it's day, and you and I weren't on one of the first wagons.
It's not that simple, unfortunately. I also have to think about having a family, and getting them a good education, so they can understand and appreciate the world. Here in the UK, and in many parts of the world, that means paying for it. Paying A LOT of money. So much that you don't see how that's possible on a normal income. So dangle a few years in finance in front of any grad, and of course they grab it.
When I was offered an internship in engineering, the pay was so low I didn't think I could rent a flat (and eat 3 meals a day). And also, the industry feels stagnant. People are always complaining about engineers not being appreciated, and British industry going down the tube. Not really something that sounds like where you want to be.
Hey, there is some sort of relative income level where I'd be happy to design stuff, but when the deals are so skewed that in a single year in finance you can make what an engineer has made his whole life, well, you have to do what's right for your descendants. Maybe they can go build stuff, and I'll appreciate that.
I'm sure science can get a bit messy, but surely what the philosophers such as Popper are addressing is a kind of boiled down process? Can you give an example? That would help a lot.
At various times people have thought there's no more to learn, and we've got it right. Even the Romans thought that. And there's a quote from around the turn of the century (1900, I forget who) saying something to that effect... that all we can do is measure the constants ever more finely. And then relativity and quantum came along not long after.
I don't know if I hope you're right or you're wrong, funnily enough.
If there were prospects for a young engineer from Oxford, maybe I wouldn't have done what most of the other engineering students did. I think it took one term before everyone realised you can work your ass off for decades designing stuff and getting paid peanuts, or you can work your ass off for a few years designing derivatives and get paid ten times that. Who in their right mind wouldn't go for the gold? That's what society is telling young engineers to go do.
True story: a derivs trader I knew was an engineer (a real one). Asked why he quit early on to work in the City: "I found out what my boss makes."
As it happens, I've carved myself a comfortable niche in the finance world, but for most people who ask me about it, I tell them it's not worth it. Long hours, lots of politics, and in the end, you'll never feel you're paid enough. And in the meantime, (if you're and engineer) you'll wonder what you could have done. My personal favourites: space ships, Formula 1, chip design.
There really are too many kids who want to work in finance. The thing is, they don't have much of a passion for finance either (it does have interesting bits, just not where everyone thinks). These kids end up screwing up both finance AND the rest of the world. Don't do it, kids.
If you mean just the number, how could someone steal the number itself? And if they did, would your car just have no number, even in databases?
They can have new plates printed. Various dealerships and auto equipment shops have machines that make plates. I'm sure a crook could get a hold of one.
I don't try to hide and lock down my car's license plate number. My car's license plate number is 6NHG617. Nobody cares about it and nobody wants to steal it. It's not valuable. The solution to the "problem" of personal identification theft is not to keep trying to hide and lock down personal information. The solution is to make personal information no longer valuable.
Are you in the UK? I went to Halford's last week, and based on my number plate, the guy at the till found out what kind of car it was, and what kinds of equipment would fit. I don't know what else he had on the screen, but I'd be pretty unhappy if it had all my details such as address, insurance details, etc. Anyway, he explained it was available as a database that firms can purchase. The fact that someone does purchase it suggests it has some value.
You don't find any use in insuring yourself against unlikely events, such as your house burning down? Surely it takes some of the fickleness out of life.
And why wouldn't rates go down? Insurers need to compete to get the business, like everyone else.
Dude, you need a new job. Other people out there will actually value your work. The thing with software is that while it's seen as absolutely critical in some organizations, others in the same field will see it as merely a utility. I work in finance, and that's how it is. Some guys want to hire some gimps to do their VBA spreadsheets. Other guys, their competitors, want to hire someone to write some proper tools that add value. Go and find them, in whatever field you're in.
I have a buddy who is an actuary. He's got great job security, insurance on his everything as a benefit, really high salary, and he gets to have long lunches around the City of London all the time. In which he can drink. I don't know why you wouldn't do it. It's even intellectually interesting, with lots of mathematical models (from early statistics to things like extreme value models). And the field itself has a history worthy of a few books (spice trade insurance, Lloyds names, suicides, asbestos, natural disasters).
Yes, some form of arbitrage has to be employed to keep the basket correctly priced. Usually this means you'll have to do it quickly, which is where certain HFT strategies come in.
But there are already markets out there where you can't see the size on the bid/offer, just current price. What do the HFTs do? Well, of course they shoot out a few market orders to gauge how much there is.
Also, if you want to eliminate the first-to-queue issue, you could just do random allocation, ie there's 100 bids, guys comes to sell 10, 10 randoms get it.
Also, in your system, is there a place for market makers? A small number of actors in the power market (where you have to trade all the time, nature of the power supply) can possibly be accommodated without (I don't know if that's what happens). But a market with large numbers and no requirement to trade might need them. And letting in MMs means letting in guys who are interested in scalping a few ticks off everyone else...
Any set of rules will have a way to play. What's the fundamental difference? The timing gap? Or the fact that the whole curve is there? People out there will spoof the curve, getting other people to think there's real demand at this or that point. And they'll guess that the cost of being called on their bluff is lower than what they make.
It basically revolves around looking at what people were doing in the market in roughly the half hour before normal market trading ended. From that, he'd make a guess (well, a calculation) as to whether people generally wanted to buy or sell. He would guess that the larger players would not manage to get their trades off, and so they would try to do the rest in the auction. He'd then trade as appropriate, and exit in the auction. (Variant of this is to enter in the auction, and exit the next morning.) Again, some kind of calculation as to where the auction would end. Doesn't sound foolproof at all, but he never lost money on it, on a daily basis. And of course I can't go into much detail, as I wasn't privileged to those details.
Any ghost hypothesis has got to have some sort of description of what a ghost is (varies across cultures) and what effects it would have on the measurable world. This could mean anything to anyone, so while you're having someone explain it to you, you might also get them to pick up the burden of proof.
Amen. You're in minority here. With me. I do what you do, and the debate in the public sphere is unbelievably uninformed.
BTW, it's not just option hedging that requires HFT. There's loads of different things to do, and some of them look silly from the outside. I've seen quite a variety of algos, all trying to do different things, on different timescales, on different markets. Of course the media tend to focus on what they've heard of, stocks.
It's not just the major banks that are in HFT. Quite a few of them are literally just guys who said "hey, why the heck can't I just rent a rack, buy some servers, and write some code?". I've met numerous guys who have done this, some as offshoots from banks, some from market making, some rather more green.
Many of the HFT firms don't even see themselves as financial firms, but rather tech firms. Latency is everything, so their conversations are about technology rather than finance.
This is more realistic, but TFA is talking about readingthe electromagnetic emissions from equipment, rather than injecting packets onto the network by ordinary means:
(A side-channel attacker looks at indirect information related to the computer -- the electromagnetic emanations from screens or keyboards, for example -- to determine what is going on in the machine. )
Slashdot Mirror
Interesting bits of finance? Well, of course it's a bit subjective. There's a lot of things that look like they're technically interesting, such as various complex derivatives, but I think the real value in those is having a sales guy reeling in the client. In fact, a HUGE number of bank/finance jobs seem to have this characteristic.
The real interesting stuff is devising your own trading strategies. You're trying to find something that orthodoxy for many years thought impossible. And most likely, anyone who found good contrary evidence didn't publish. So it's pretty thin what you have to start with; a field where your own discoveries are actually your own, even though probably other people have come up with similar ideas (and aren't telling). Having no orthodoxy is great for seeing whether you actually understand how statistics work, and in the wider picture, how one discovers anything.
By the way, with your background, there's a good chance you'll be shoehorned into some derivs type thing. You'll definitely be able to do it, although there's some new math. It's initially interesting (pick up the autobiography of Emmanuel Derman), and definitely pays well, but ultimately loses it's lustre. I think the gold rush from science to Wall Street has had it's day, and you and I weren't on one of the first wagons.
Eh, don't feel so bad. The ones I've met have tended to be very warm, whether they were intelligent or stupid.
He died in 1945...
This is very logical, and is exactly what people think they will be doing...
It's not that simple, unfortunately. I also have to think about having a family, and getting them a good education, so they can understand and appreciate the world. Here in the UK, and in many parts of the world, that means paying for it. Paying A LOT of money. So much that you don't see how that's possible on a normal income. So dangle a few years in finance in front of any grad, and of course they grab it.
When I was offered an internship in engineering, the pay was so low I didn't think I could rent a flat (and eat 3 meals a day). And also, the industry feels stagnant. People are always complaining about engineers not being appreciated, and British industry going down the tube. Not really something that sounds like where you want to be.
Hey, there is some sort of relative income level where I'd be happy to design stuff, but when the deals are so skewed that in a single year in finance you can make what an engineer has made his whole life, well, you have to do what's right for your descendants. Maybe they can go build stuff, and I'll appreciate that.
I'm sure science can get a bit messy, but surely what the philosophers such as Popper are addressing is a kind of boiled down process? Can you give an example? That would help a lot.
At various times people have thought there's no more to learn, and we've got it right. Even the Romans thought that. And there's a quote from around the turn of the century (1900, I forget who) saying something to that effect... that all we can do is measure the constants ever more finely. And then relativity and quantum came along not long after.
I don't know if I hope you're right or you're wrong, funnily enough.
Even I think that, and I work in finance.
If there were prospects for a young engineer from Oxford, maybe I wouldn't have done what most of the other engineering students did. I think it took one term before everyone realised you can work your ass off for decades designing stuff and getting paid peanuts, or you can work your ass off for a few years designing derivatives and get paid ten times that. Who in their right mind wouldn't go for the gold? That's what society is telling young engineers to go do.
True story: a derivs trader I knew was an engineer (a real one). Asked why he quit early on to work in the City: "I found out what my boss makes."
As it happens, I've carved myself a comfortable niche in the finance world, but for most people who ask me about it, I tell them it's not worth it. Long hours, lots of politics, and in the end, you'll never feel you're paid enough. And in the meantime, (if you're and engineer) you'll wonder what you could have done. My personal favourites: space ships, Formula 1, chip design.
There really are too many kids who want to work in finance. The thing is, they don't have much of a passion for finance either (it does have interesting bits, just not where everyone thinks). These kids end up screwing up both finance AND the rest of the world. Don't do it, kids.
If you mean just the number, how could someone steal the number itself? And if they did, would your car just have no number, even in databases?
They can have new plates printed. Various dealerships and auto equipment shops have machines that make plates. I'm sure a crook could get a hold of one.
I don't try to hide and lock down my car's license plate number. My car's license plate number is 6NHG617. Nobody cares about it and nobody wants to steal it. It's not valuable. The solution to the "problem" of personal identification theft is not to keep trying to hide and lock down personal information. The solution is to make personal information no longer valuable.
Are you in the UK? I went to Halford's last week, and based on my number plate, the guy at the till found out what kind of car it was, and what kinds of equipment would fit. I don't know what else he had on the screen, but I'd be pretty unhappy if it had all my details such as address, insurance details, etc. Anyway, he explained it was available as a database that firms can purchase. The fact that someone does purchase it suggests it has some value.
You don't find any use in insuring yourself against unlikely events, such as your house burning down? Surely it takes some of the fickleness out of life.
And why wouldn't rates go down? Insurers need to compete to get the business, like everyone else.
Dude, you need a new job. Other people out there will actually value your work. The thing with software is that while it's seen as absolutely critical in some organizations, others in the same field will see it as merely a utility. I work in finance, and that's how it is. Some guys want to hire some gimps to do their VBA spreadsheets. Other guys, their competitors, want to hire someone to write some proper tools that add value. Go and find them, in whatever field you're in.
I have a buddy who is an actuary. He's got great job security, insurance on his everything as a benefit, really high salary, and he gets to have long lunches around the City of London all the time. In which he can drink. I don't know why you wouldn't do it. It's even intellectually interesting, with lots of mathematical models (from early statistics to things like extreme value models). And the field itself has a history worthy of a few books (spice trade insurance, Lloyds names, suicides, asbestos, natural disasters).
I don't get the hostility...
I don't get it, why the hostility? All they do is calculate risk for insurance purposes.
Yes, some form of arbitrage has to be employed to keep the basket correctly priced. Usually this means you'll have to do it quickly, which is where certain HFT strategies come in.
I actually like your idea quite a bit.
But there are already markets out there where you can't see the size on the bid/offer, just current price. What do the HFTs do? Well, of course they shoot out a few market orders to gauge how much there is.
Also, if you want to eliminate the first-to-queue issue, you could just do random allocation, ie there's 100 bids, guys comes to sell 10, 10 randoms get it.
Also, in your system, is there a place for market makers? A small number of actors in the power market (where you have to trade all the time, nature of the power supply) can possibly be accommodated without (I don't know if that's what happens). But a market with large numbers and no requirement to trade might need them. And letting in MMs means letting in guys who are interested in scalping a few ticks off everyone else...
Any set of rules will have a way to play. What's the fundamental difference? The timing gap? Or the fact that the whole curve is there? People out there will spoof the curve, getting other people to think there's real demand at this or that point. And they'll guess that the cost of being called on their bluff is lower than what they make.
Interesting idea however.
Dude, that's what people do. Well, some people. BTW, how exactly do you enter your whole supply/demand curve? How do you keep it honest?
It basically revolves around looking at what people were doing in the market in roughly the half hour before normal market trading ended. From that, he'd make a guess (well, a calculation) as to whether people generally wanted to buy or sell. He would guess that the larger players would not manage to get their trades off, and so they would try to do the rest in the auction. He'd then trade as appropriate, and exit in the auction. (Variant of this is to enter in the auction, and exit the next morning.) Again, some kind of calculation as to where the auction would end. Doesn't sound foolproof at all, but he never lost money on it, on a daily basis. And of course I can't go into much detail, as I wasn't privileged to those details.
That's not true of all auctions. Certain markets show you the current clearing price only. But it's been a while since I looked at them.
OMG. You've just reinvented the auction. And guess what, I know a guy who writes algos to game this exact thing.
Any ghost hypothesis has got to have some sort of description of what a ghost is (varies across cultures) and what effects it would have on the measurable world. This could mean anything to anyone, so while you're having someone explain it to you, you might also get them to pick up the burden of proof.
Amen. You're in minority here. With me. I do what you do, and the debate in the public sphere is unbelievably uninformed.
BTW, it's not just option hedging that requires HFT. There's loads of different things to do, and some of them look silly from the outside. I've seen quite a variety of algos, all trying to do different things, on different timescales, on different markets. Of course the media tend to focus on what they've heard of, stocks.
It's not just the major banks that are in HFT. Quite a few of them are literally just guys who said "hey, why the heck can't I just rent a rack, buy some servers, and write some code?". I've met numerous guys who have done this, some as offshoots from banks, some from market making, some rather more green.
Many of the HFT firms don't even see themselves as financial firms, but rather tech firms. Latency is everything, so their conversations are about technology rather than finance.
This is more realistic, but TFA is talking about readingthe electromagnetic emissions from equipment, rather than injecting packets onto the network by ordinary means:
(A side-channel attacker looks at indirect information related to the computer -- the electromagnetic emanations from screens or keyboards, for example -- to determine what is going on in the machine. )