Career in finance

chill_factor

self delete off topic post.

mdxyz

It doesn't. Although you can see here too that the improvements from "really really bad" to "really bad" have been good enough to net 8% growth. India is ranked lower. It didn't copy the British. Post-independence India is actually a very interesting country, because it adopted the Soviet/PRC economic model while remaining a democracy with many of the British-era social freedoms. As it turns out, this didn't work better at producing GDP, though far fewer people ended up as victims of political murders and enslavement.

Adopting market economics would solve the more egregious problems (starvation, etc.) of every country, yes.

edit: the above post is somewhat different to what it was when I replied.

chill_factor

self delete off topic post.

mdxyz

It's misleading to view there as being this "you" which is a single unified embodiment of a country, and of other countries. There's no difference in principle between a US company building a railway in a different part of the US, and a US company building a railway in Burundi, or a Burmese company building a railway in Guatemala. North Dakota isn't 'economically independent' but it isn't being 'held hostage' by California. Nor is any sufficiently small locality able to do almost any of that stuff for itself. In developed countries you get the opposite of 'economic independence': everyone is highly specialised, and as a result couldn't survive without everyone else, but it's actually better because they are able to perform their narrow specialism much more productively, and so is everyone else.

So why are there major differences in economic outcome across national borders? The bottom line is that in some countries it is illegal to make investments that would yield a return at the market rate, or the government would take too much for it to be worthwhile, or the other problem, which they have in Somalia, where the law is so weak that some random guy might come and kill you and take it. A market economy is just one where it's generally speaking legal to trade across borders, to trade with other individuals within your borders, and where some money you have today, you will still own next year.

The state could in principle do the same things. The problem is that it doesn't work in practice, and there are good theoretical economic reasons for that.

I'm not sure. I find that a lot of physicists don't know much about economics and this gives them funny ideas about how the economy does/should work. That is very relevant for careers.

chill_factor

self delete off topic post

mdxyz

Not really. Solid evidence of the 20th century has been that market economies work, and command economies don't. The baffling thing is why people kept trying command economies despite all the failures, and even more so, why a lot of people still favour them today.

Devils

“There comes a time when you ought to start doing what you want. Take a job that you love. You will jump out of bed in the morning. I think you are out of your mind if you keep taking jobs that you don't like because you think it will look good on your resume. Isn't that a little like saving up sex for your old age?” - Warren Buffett

Real businesses generate economic value. All that the banks have done to the US in the past 5 years is nearly send it bankrupt. Sure you can make huge money, but you can lose money too. IQ is correlated with income but not wealth http://thesocietypages.org/socimages...me-and-wealth/ . Using your intelligence to make money is easy, becoming wealthy is an art.

To inject some realpolitik, there is another type of "investing" too - running your own small business where 25% return on capital is not unusual. I've made money by running an IT consulting business. I was reading about somebody with 2 hotdog stands in New York making $120K, hard work and common sense is probably more important then IQ in making money. "The Millionare Next Door" has good insights on how average people can get wealthy, we dont all want to live in New York doing program trading http://www.amazon.com/Millionaire-Ne.../dp/0671015206 . And boy have I made some mistakes running my own businesses but I have learned a lot too, and you eventually get to the "lightbulb moment" of knowing how investing & running businesses really works, & to save money, invest wisely, and prosper.


My point is do when you like, have a background plan to say make $1 million in 20 years. Do what you like in your 20s and aim to have $1 million by age 40 as your worst case scenario. You can be in the job you love and make money on the side as a hobby like I have done.

Devils

I have a math degree & have tinkered around with mathematical finance out of interest. Having formally studied real analysis, complex analysis, numerical methods, engineering math, linear algebra etc then mathematical finance is just another textbook application to me. As a math graduate you get a "feel" for the functions, an armory your have learned in solving the math, built up from solving thousands of other math problems.

So when out of interest I wanted to develop personal software for delta neutral trading http://www.optiontradingpedia.com/de...al_trading.htm I did it using knowledge of partial differential equations & numerical methods I have used in many other situations.

The same isnt true for somebody who "wants to be a quant", doesnt have a math background and uses rote learning to understand the math. Sure you learn the equations but without the deeper understanding a mathematician has.

DeepThoughts

Well guys, I am a finance major. I grew up with finance people. And I still talk with them about this same concept. The problem is your trying to chase the golden goose and EVERYONE is trying to do the same. The hedge fund count has exploded and so have the number of people with physics backgrounds trying to do the same thing as you. In fact, there are university programs now at schools like NYU that are tailoring programs in order to get the exact blend your talking about. These "quant farms" are already at the top of the trend and your competition is going to be just the same. Not that what your doing is impossible, it just has lost its nitch.
As far as trying to get the best education for the demand of the future? No one knows. Thats the next golden goose. And if everyone knew how to get there it wouldn't be the golden goose!
Most of what your talking about is number crunching and patterns. Assuming your going for short-tearm high-frequency investing what you need to be able to do is take a load of data. Find a general pattern. And prey to god the market conditions hold and the computer program has the right parameters.
People in this field also have a tendency to work very long hours. If you dont love the realm of long team unpredictable crap(longer than a week) like I do. I wouldn't sacrifice a solid steady career in physics. I know a nuclear engineer that makes USD$200,000+ a year and works 35 hours a week. Not bad. So the question then becomes. Whats ANY job that meets your Time to Pay equilibrium. I hate to be a naysayer but WallSt. realm is a total meet grinder. It will take you for all your worth and thats it. I would get the PHD as a back up!

twofish-quant

That's wrong.

University researchers can and do have access to lots of non-public information which they can and should filter for public consumption. Let me give you a real example. I can't post here what the research that I'm doing, because I don't know that someone won't take that information and try to make money off the stock market. Even if it doesn't happen, the SEC will can get annoyed if I say something that can be interpreted as promoting a stock.

Now, what I *can* (and in fact do) do is to talk with people in local universities about what we are working on. I can do this because I know and trust the person that I'm talking to is not going to buy stocks with that information, and from the point of view of the SEC, I'm not making a "public offering." The reason I do this is so that the professor can then take that information and then adjust their course offerings to reflect what I've told them.

There is a lot of mathematical research that we are doing that we can't make publicly available. However, we do have a set of academics that are "white gloves." We feed them our mathematical research, they make some adjustments, and then they publish. Our name is not on the papers, and we and the regulators prefer it that way.

But how do you know that you've got it right?

1) One thing that's fun is that often people don't know what they don't know.

2) I don't think that bad advice is the big problem. The bigger problem involves cultural biases. Yes you *could* go into industry, but you are a rotten nasty loser if you do it.

3) One thing that I found was that getting the skills that I needed to be productive in industry made me much less competitive when I applied for graduate schools. Because I was reading marketing books, I got "B"'s in classes that I could have gotten A's in, and that seriously hurt me when it came time go get into graduate school.

I think that assumes that people are more autonomous than they actually are. One thing that I've found interesting is that one big trick that sales people do is to give people the illusion of choice. Sales people can't point a gun at your head and force you to buy a car, but they have a lot of psychological tricks that they use to push you in that direction.

Non-profit universities do not pay tax. If the universities owns a for-profit business that's taxable.

But by working for the public good, you are subject to higher standards. If I go to buy a car, I know that the salesman is not acting in my interests, and if the salesman lies to me and convinces me to buy the car, to some extent, it's my fault since I should have known not to trust a car salesman completely.

Because universities are supposed to work for the public good, there are higher standards. The difficulty is that universities have "private interests." Universities are businesses and they are under pressure to increase tuition. If you go to a car salesman, and ask "do I really need this car?" the salesman will say *absolutely* and then pull out every psychological trick that they can to get you to sign on the dotted line.

The difficulty for universities is that they are also under similar pressures, but the rules are different for public institutions. That means that if the provided bad advice in the past, you just can't say "your fault." You can't change the past, but having people say "I'm sorry we screwed up, what can we do to make it not happen again?" would help. I don't hear people saying that.

1) It's a critical factor

2) University budgets are the responsibility of the individual states, but most of the money is Federal. State constitutions mandate balanced budgets, but there is this big loophole that states receive Federal transfer payments, and if you look at where the money ultimately comes from, it's mostly Federal.

Well, yeah.......

But what you can do is to hold enough protests at the state level to stop cuts, that the state legislatures are forced to beg for money from the Federal government. If the state legislatures can balance their budgets by cutting state universities, they will. Whoever screams least gets cut. If you scream at the state legislatures not to do that, then the numbers won't match, so they'll have to go to the Federal government. At which point you play the game in Congress. The Federal government can do two things that state governments can't. It can run deficits and print money.

One person can't do much, but one person in an organized system can do a lot.

Which are being cut by state governments. One good thing about community colleges is that getting a two-year degree from a CC is often a stepping stone to a four year degree at a local public university.

Now what I'd like to see is to see an alliance between MIT and community colleges. Personally, I think it's a mistake for MIT to focus on creating a distance education partnership with Harvard. The institution that MIT really should reach out to is Roxbury Community College. One thing that RCC can provide is vocational training so that physics Ph.D.'s don't starve once they get their degree.

But it does. If you know someone that works in a major bank, they can give you the e-mail of someone that you can send your resume to with a reasonable assurance that someone will at least read the resume. If you hang around managers, HR people, and CEO's, you figure out how they think and that gets you in the door.

The trouble is that when there is a high unemployment rate, the knives come out. If there are N job openings and 1.1 N job seekers then increasing training is just going to make the problem worse.

One problem with increasing training is that you end up with the Ph.D. problem. If we train 100,000 air conditioning repair men then you will likely end up with 90,000 unemployed air conditioning repair men who are worse off because they have debt.

[QOTE]You are assuming here that China will continue to maintain their impressive rates of economic growth, which is far from guaranteed[/QUOTE]

People in China are looking ahead and seeing problems and they are dealing with them. There are limits to China's current growth model, and it is going to hit them in the next few years. That's why people are working *NOW* to start fixing the problems so that you don't end up with an economic crisis in ten to twenty years.

The general belief in China is that long term, China needs to move into science and technology. This is going to take two to three decades to do, which is why people are starting *NOW*.

The thing about China is that it has a plan. It may be a bad plan, but people are thinking about what the Chinese economy is going to look like in 2050, and it's going to have a lot of science and technology. What worries me about the US, is that no one is even thinking about 2050. I haven't even heard people talk about 2020. This is bad for science. If your time horizon is two years, then science is a waste of money. If your time horizon is fifty years, then it's stupid to cut science now.

The good news for the US is that it has such a good science infrastructure that it's going to take a decade of incompetence to destroy it. People are being stupid now, but there is still lots of time to stop it before things get bad.

So rather than sit back and let things fall apart, the government is actually *doing stuff*. If you have lots of unemployed people, they start getting angry and demonstrating. So what can we get them to do to avoid that. Well, there are some railroads to be built.

Employment is essential to keep the government in power, so the government is moving heaven and earth to get people jobs.

So maybe it's not such a good idea to move to a consumer-based economy. Singapore has high savings rates, and if through science and technology you can maintain high growth rates, you can keep saving massive amounts of money.

Savings -> science -> technology -> productivity -> more savings

The big limit is when people get old and start withdrawing their savings. How do you keep productivity with fewer workers. Science to the rescue!!!!

Sometimes you don't. There's something to be said for saying "we are angry, we just want results, you are the people in power, you figure out how to get us stuff." Getting people to think too much is "psychology trick #2343" for making sure that nothing happens.

twofish-quant

With enough time, you can make large institutional changes. For a developed economy, 3% is outstanding. The question for a developed economy is whether you want 0% without technology or 3% with technology. Historically three percent growth is huge. That was the average growth rate of the industrial revolution.

Over a lifetime, 1% growth means 2x wealth, whereas 3% growth means 8x. Chinese growth rates are going down, and the question is whether they are going to stabilize at 3% or 0%, and to get 3%, you need science and technology.

Does it really matter?

The trouble with getting into arguments over what is "best" is that it's an excuse to do nothing.

BTW, I've read Von Mises, and know about the socialist calculation problem. I actually agree with a lot of stuff that he says, but I think he gets some things critically wrong.

There are some things that governments are horrible at. Some things that governments are good at. In any event, the government in 2007 realized that if it didn't generate employment and fast, it would be faced with what happened in Egypt, so the government went and asked what project would keep people too busy to demonstrate, and if it produces economic value, that's a nice side effect.

No, they move people from "unemployed" to "working." It may be acceptable for the private sector to have lots of people doing nothing, but in China having lots of unemployed people means mass protests, and so it's unacceptable. One reason that the US has higher unemployment than China is that US political system can tolerate 8% unemployment indefinitely without a revolution. China can't.

The thing about survival is that it's not good enough for most people. No one in the US need to starve and the biggest problem with poor people is obesity and not malnutrition. If you like the way things are, and don't see any need for making things better, faster, cheaper then there really isn't that much need for science.

The thing about China is that it's no longer in starvation mode. The last famine in China was in 1973. So if Chinese people were merely content with "not starving" then you don't need to do anything else. But people aren't content with that. People want more stuff, and for that you need constant economic growth.

China's current growth model is going to hit a wall around 2020. Since the government wants to stay in power past 2020, people are trying to figure out what is going to work. Part of the reason people are worried NOW is that the first plans are going to fail.

Just to be clear, I'm not worried about the United States collapsing economically. What I am worried about is that the US turns into Japan, and loses the "frontier spirit".

twofish-quant

I think it's a *lot* more complicated than that. One problem with "market rate" is that you have to have a market, and getting one is non-trivial. One problem is that people sort of assume that if you don't have a "free market" system then you have a Stalinist system when in fact there are other possibilities.

I find that most theoretical economists just end up formalize how they think the world should work rather than looking at how it really does work.

A "hands off" policy toward the banking system also doesn't work. The trouble is that the state is the ultimate guarantor of the banking system, so if the state isn't heavily involved at regulating the system, what people will do is to make "head I win, tails the state loses" bets.

Now in principle, you could have things work with a "no bailout" policy. The trouble with that is that you can set up a situation in which "no bailout=the world blows up". At that point the state is forced to do a bailout. The trouble is that if people know that the state is going to to a bailout you get back to the previous situation. You *say* that you aren't going to bail me out, but I don't believe you.

twofish-quant

Not so clear. The Soviet economy seemed to work very, very well in the 1950's. It fell apart around 1970. Now *why* it fell apart is an interesting question. I would argue for Vladimir Popov's model in which the Soviet Union stopped infrastructure investment in the 1950's.

Essentially, the trouble was that because Soviet planning was based on production quotas rather than profit, there was strong disincentive to replacing aging machinery. You lost your quota, and there was no way of calculating the economic benefit of the new machines. This didn't matter in 1950, because the Soviets had to repair everything because of WWII, but it started to kill the economy in the 1970's. If that's the explanation, then it's going to be pretty bad for the US (however, unlike the Soviet Union, the US has a strong enough political system to prevent total collapse).

Because no one arguing for a Soviet command economy. One thing about China was that since 1950, "Soviet=bad" so that in the late 1970's, there was strong resistance to reinstituting Soviet style economic planning. One thing about Soviet economic planning is that it requires something like a Josef Stalin, and the Communist Party in 1980 (or now) simply does not have that sort of power or economic unity.

Having some central planner order a factory to produce 500 locomotives just will not work well for the reasons that von Mises and Kornai pointed out. Now what I think will work is to have a pricing system in which you reflect economic costs. So a government bureaucrat looks at the cost of 500 locomotives and can compare that with the price of say 10 aircraft carriers. By changing tax rates and purchasing things at market prices rather than forcing people to build them, you get around the socialist calculation problem.

There are some similarities between the Chinese economy and the theories of Oskar Lange. It's not a concidence since there was a lot of Eastern European influence in Chinese thinking circa 1980.

I used to wonder why people clung to Soviet models after it seemed obvious that they wouldn't work. Then I see people cling to neo-liberal economic ideas after it is pretty clear at least to me that there is something seriously wrong with them, and it makes more sense. One issue is that you can't replay history, so when the Soviet model fails, you can try to argue that there wasn't enough "real communism", and I've seen similar arguments that things fell apart because we don't have "real markets." You can get into large arguments over this, but the counterargument that I find convincing is a political one. If you can't realistically have "perfect communism" then it's pointless to argue about whether it is a good thing. Similarly if you can't have "perfect capitalism" then it's pointless to argue that it's a good thing.

One problem is the idea that if you oppose X you must support Y. I think that Soviet economic planning doesn't work, but Chinese style economic planning works in China (whether it would work or not in the US, I don't know, and I doubt it).

twofish-quant

Astrophysical simulations use a lot more compute power than most engineering simulations. Also most engineering simulations can be done with "canned programs". Essentially the equations are the same from situation to situation whereas the boundary conditions are different.

In oil/gas and finance it's the opposite. The boundary conditions are trivial whereas the dynamics changes.

atyy

How can Singapore be more of a "free market" when a majority of its population lives in (really good) government housing?

twofish-quant

So how did Russia win? One reason people in China were attracted to the Soviet model in the 1950's was watching the Russians just annihilate the Japanese army. Gee. Wish we could do that......

Don't think so. Aircraft was invented in 1903, and you just can't copy an aircraft carrier or build aircraft factories. It's actually both sad and enlightening to see the problems China has now trying to build civil aircraft and aircraft carriers. It's pretty clear that Chinese civil aircraft manufacturers and military shipyards are pretty clueless. But if you don't try to build a rotten airplane, you'll never build a good one.

There's also the little matter of whether other countries will let you copy their stuff. I'm sure that China would love to have the latest designs for American aircraft carriers. One big reason China is working on high-speed rail rather than aircraft is that most countries are much less annoyed when you copy their railroad designs than if you copy aircraft designs, because there are fewer military applications.

Because of Russia. Also, if you are in a life and death struggle for national survival, you can make the argument that command economies are good. Militaries are not run on market principles, and one way of thinking of China in 1940 was to imagine the entire population being one big army. This also happened in the United States. Once you are on a war footing, then you have price controls and rationing.

Now you can argue that people in 1950 were mistaken, because once Russia got to 1970, things fell apart. But one wonders if the same thing is happening to the "capitalist model". I can imagine someone in 2020 looking at the "neo-liberalism" the same way that people in 1995 looks at communism.

There's also generational issues. The Cold War was an important part of my childhood, and the fall of the Soviet Union was one of the important defining events of my youth. So it's interesting to know that people in college today didn't experience that and their view of the world is likely to be quite different because there are things in my memory that aren't in theirs.

twofish-quant

So does much of Hong Kong. Hong Kong is "central planning done right." The HK leaves some things to the market, but it is extremely intrusive in other things (such as health care where HK copied the British model). There are some areas (land policy) where you see the PRC copying Hong Kong.

The Singapore government is also pretty intrusive with things like Temasek holdings.