Is it possible to calculate the prices of stock market trends with calculus?

[QuantumTheory]

My grandpa is an investor . I heard some people get rich off the stock market.

Is it possible to predict stock market prices?

thank you

 

[Pengwuino]

It is not possible to predict stock market prices and it is DEFINITELY not possible to calculate them using calculus.

The problem with mathematically deducing what a stock will do is that the stock market does not follow equations or anything like that. The "variables" in the stock market are beliefs, feelings, instincts, etc instead of ratios and constants and such. It is kind of like asking if you can mathematically predict love using calculus. Is there an equation that will tell you if someone will fall in love with you? No. Same with the stock market.

People get rich off the stock market for various reasons. An extremely extremely extremely small # of people get lucky and throw money into a stock and become rich because the stock just happens to take off. A majority of people however, get out after losing or gaining a few dollars/share. A lot of people throw their money in and the stock falls and they just think its going to go back up one day but it never does and they lose a large portion of their money. A small number of people center their lives around the market however. They look at tremendous amounts of information related to specific stocks they are researching to determine how the stock might turn out and make various judgement calls and act on those calls. Those are the people who really get rich off the stock market.

Another group of people that gets rich however, are corporation presidents/CEO's. For example, we'll use Bill Gates. The reason he is rich is because Microsoft was his company. Once they went public and offered shares, Bill Gates had a large portion of the shares of the company; a large % to be exact. As Microsoft grew to have a networth higher then many countries do, Gates effectively became worth the $50 billion or whatever he's worth now (or is it 10, i dunno). This is because say, if he had stock of 50% of the company and the company was worth $100 billion, he is effectively worth $50 billion. The thing is however... he doesn't really have this cash accessible. He would have to sell his stock which means he sells off portions of his ownership in the company.

There is one good example of what the real stock market works off of, beliefs. If Bill Gates started selling off large # of stocks, people will start believing and speculating that something is wrong with the company. This will mean that the stock share will command a lower price on wall street because it is no longer as good of a company as people think it is. Conversely, if some big wall street brokerage starts buying up shares of Microsoft, people will start thinking Microsoft is becoming a better company and closer to achieving its goals of world-wide domination. Thus, a good stock to get into.

Another implication of this system is that brokerage houses can literally destroy companies or pull failing companies out of the grave. For example, I believe Krispee Kreme went through this. Brokerage houses and investors had large amounts of money in the company (I may be wrong as to which company I am thinking of though). They said they were going to expand and everything was going to be great. Well they hit the west coast and they weren't successful. The investors, who held tremendous amounts of stock in KK, realized the company wasn't worth it so they all sold off. KK's stock price then tumbled. For many companies, this action could very well destroy them. Converseley, a company can be pulled out of the ground because of a large brokerage house or large investors. I can't really think of an example right now but what basically might happen is that a brokerage house might start buying up shares of a company like crazy. This would mean the company has a larger networth now. They could also see this and issue more stock and raise capital to expand the company. Since the brokerages think its a good company, they can gain a large amount of money through stock issuance (the company that is).

 

[FredGarvin]

Don't you think if it were possible, there would be some very very wealthy people out there?

 

[Anttech]

Most brokerages are using Mathimatical alogriths to help them deside when to invest and when to withdraw stocks...Also when assembeling your portfolio mathimatics can help you minimise risk on a certain gain. Investors also look at GARP to determine what is a wise investment and what isnt.

In recent years as computers have been used to do a lot of number crunching, many "quants," as they like to call themselves, have gone completely native and will only buy and sell companies on a purely quantitative basis, without regard for the actual business or the current valuation - a radical departure from fundamental analysis. "Quants" will often mix in ideas like a stock's relative strength, a measure of how well the stock has performed relative to the market as a whole. Many investors believe that if they just find the right kinds of numbers, they can always find winning investments. D. E. Shaw is widely viewed as the current King of the Quants, using sophisticated mathematical algorithms to find minute price discrepancies in the markets. His partnership sometimes accounts for as much as 50% of the trading volume on the New York Stock Exchange in a single day.

https://db.tradek.com/w/content/howto/trading/trading_ib6.html

 

[quantumdude]

Don't you think if it were possible, there would be some very very wealthy people out there?

Yeah, jeez, if it were possible to do that, I wouldn't be sitting around here talking to you people.

 

[Lisa!]

Yeah, jeez, if it were possible to do that, I wouldn't be sitting around here talking to you people.

We had a very knowledgeable professor who hardly spent time to answer students' questions out of the class( I was an exception of course ). he was always sitting in front of his computer and we thought he was sutudying physics articles, but once 1 of his student had discovered that actually he was visiting a stock market site! And I should tell you he was really rich ,although he was only 35! So yes, it might be possible and you...

 

[vanesch]

Is it possible to predict stock market prices?

I know some people who TRY to do this. In fact, this business used to be an attraction pole for physicists-out-of-their-25th-postdoc. Of course it is difficult to predict individual stock prices, but they try to find some correlations between stocks in order to be able to predict short-term moves. I knew some people (I don't know if they were very successful) who just tried to fit big neural networks onto a lot of stock data, with the hope that the neural network would fit onto the correlations. But of course, major events are never included into this. The result of elections or so cannot be foreseen. You can only hope to find some short range correlated dynamics (if the course of oil companies goes up, then maybe a few days later, the course of car manufacturers goes down or something like that). That can be enough to make small margins, and if you have enough money to bet onto your model, you can make bucks with it.

However, another known truth is that such a model cannot be generally known. If it were, the market would ALREADY correct for it by all those greedy nerds all over using their model, and as such, no margin would be left.

 

[Physics_wiz]

I read somewhere that stock prices are chaotic, which implies that they are very sensitive to initial conditions. If anyone figured out a pattern and tried to draw a profit out of it, they would fail because they would change the initial conditions.

 

[QuantumTheory]

Thanks for the great post Pengwuino
PHysics whiz, do you know a lot about physics?

I'm 17 and don't know what to do yet, and I hope this will help me decide,thanks!

Pengwuino I will ask you more questions on stock by PM, is that ok with you?

 

[Physics_wiz]

Thanks for the great post Pengwuino
PHysics whiz, do you know a lot about physics?

I'm 17 and don't know what to do yet, and I hope this will help me decide,thanks!

Pengwuino I will ask you more questions on stock by PM, is that ok with you?

It looks like we have some stuff in common . Well ok, first, I don't really know much about physics. I created this account when I was in high school and was pretty good at my AP physics class. Right now, I'm in engineering...still know some physics but not much. I recently got interested in stocks and the stock market too and got a few books. I even decided on buying a few SPY shares for now until I read up more on the stock market. If you have stock questions, post them here! I'd be happy to discuss them or learn something new from the thread. Guess what? When I first started, I downloaded a few spreadsheets of data to check for any trends...I was a bit stupid. Anyways, read a few books on the stock market, they help a lot. Also, I found "Fooled by Randomness" to be an awesome book. It won't teach you much, but the author makes some very good points (pay attention to what he says) that most 17 years old or 19 years old (me) people don't know about.

 

[fourier jr]

Don't you think if it were possible, there would be some very very wealthy people out there?

ya if it were possible to use calculus to get rich off the stockmarket wouldn't literally everybody who follows that stuff make a killing? it's like asking if statisticians & probabilists know any tricks to use at a casino.

 

[QuantumTheory]

It looks like we have some stuff in common . Well ok, first, I don't really know much about physics. I created this account when I was in high school and was pretty good at my AP physics class. Right now, I'm in engineering...still know some physics but not much. I recently got interested in stocks and the stock market too and got a few books. I even decided on buying a few SPY shares for now until I read up more on the stock market. If you have stock questions, post them here! I'd be happy to discuss them or learn something new from the thread. Guess what? When I first started, I downloaded a few spreadsheets of data to check for any trends...I was a bit stupid. Anyways, read a few books on the stock market, they help a lot. Also, I found "Fooled by Randomness" to be an awesome book. It won't teach you much, but the author makes some very good points (pay attention to what he says) that most 17 years old or 19 years old (me) people don't know about.

Heh I am interested in a lot of things right now. mostly numismatics, and physics (time dilation and general and special relativity) its one of those things i just want to know without having to understand all the complex equations (if you think this is possible, let me know) and economics. Are there any jobs you can do with a degree in economics? do you know of any good books on stocks for beginners?
thanks

 

[Pengwuino]

Pengwuino I will ask you more questions on stock by PM, is that ok with you?

Sure, i'll try to do my best to not screw up your understanding :D

 

[Pengwuino]

ya if it were possible to use calculus to get rich off the stockmarket wouldn't literally everybody who follows that stuff make a killing? it's like asking if statisticians & probabilists know any tricks to use at a casino.

But there is tricks to use at the casino!

Except that its a bit difficult and casinos look for people who are doing it and kick them out.

 

[cronxeh]

There are a few tools like Stochastic models and derivative pricing, Ito calculus, Black-Scholes partial diff eq model, Heath-Jarrow-Morton (HJM), Hull-White (HW), Monte Carlo, Black-Karasinski (BK), Cox-Ross-Rubinstein (CRR) for option pricing, Equal Probabilities Tree, Crank-Nicolson, Black-Derman-Toy (BDT), Brace-Gatarek-Musiela, Ho-Lee

 

[fuzzyfelt]

Hi Guys - this is Husband Of Fi: darling wife brought this thread to my attention since i have only ever worked in stock markets, and that's for 15 years now.
I think the biggest issue for modelling markets is the almost infinite number of variables (each participant is a variable in themselves!). Somebody with out-math me here i am sure, but there is a big cause and effect problem with economics, ecascerbated by the fact the most of the variables are organic and emotive rather than rational. Thus, its always been my view that the best discipline to possesses is psychology rather than economics. Now i, like many of you perhaps, would have thought that human nature (being the driver) was modelable, but long story short, its not! There are a few "natural number" (what is the term??) rules that certainly do get you somewhat ahead of the pack, eg, Fibonacci levels, Stochastic Oscillators, but the bulk of the analysis we use is of a statistical nature. The bulk of predictive modelling as traded by traders (eg hedge funds, rather than long only investors, eg Franklin, Fidelity etc) is one way or another based around mean reversion, this is a better than average sort of modelling too, and somewhat organic in its origins. The best rules are the simple ones, and the emotive ones, the market is driven by only two basic principles: Fear and Greed. I think for markets a 5 year view isn't too hard (thats macro economic stuff) and a 5 min view isn't so hard (thats trend and momentum) the tough views, and the ones that really pay to get right, are the 6 mth and 1 year views: splitting a chicken is probably as valid a method as any for that!
The interesting thing about bubbles is - everyone knows its a bubble, therefore the only value is picking the catalyst for the correction, not seeing the bubble itself! In the market there is intellectual snobbery associated with being bearish, becuase the great unwashed masses are always bullish and optimistic - i enjoy the irony of the fact that all charts basically start in the bottom left and corner and finish in the top right: therefore, the unwashed masses ploughs over the intellectual bears every time! yay the common man!
Re the question on books: a couple that i and a few others i know in the market have enjoyed; Market Wizards, there was 1 and 2 at least, this is a book of traders anecdotes as is interesting, and, Irrational Exhuberance by Robert Schiller which is an interesting look at late 1990s equity valuations.
Get into the markets, its fun! and even more fun with somebody elses money!

 

[math dunce]

I came across this question by accident, and although I agree that you probably couldn't be accurate 100% of the time, you could use math to outperform the market. Before I start, I have to admit that I am mathematically challenged. That said, I believe that by picking a small group of stocks (5-25) and assigning numeric value to certain attributes, you would increase your chances of being profitable. I trade stocks daily and average 5% a month. It won't make you a millionaire overnight but you should be able to double your money every 14.4 months. I do this without any algorithms, but, I think if you used math you could make a fortune! At the very least you would lower the risk that the market would slaughter you. I would do it but I can only do so much with my abacas.

 

[elfboy]

I have done work on this this for awhile and have derived equations that describe the mechanics stocks market based on empirical observations. it works by breaking buy & sell orders into individual 'electrons' and 'positrons' and then there are other variables like volume

 

[cronxeh]

god damnit.

 

[edward]

THE ALCHEMISTS OF WALL STREET

An in depth look at Quants. Don't be fooled by the beginning the man is a multi-millionaire.

 

[Loren Booda]

Try catastrophe theory (aptly named).

http://en.wikipedia.org/wiki/Catastrophe_theory"

 

[Gokul43201]

I have done work on this this for awhile and have derived equations that describe the mechanics stocks market based on empirical observations. it works by breaking buy & sell orders into individual 'electrons' and 'positrons' and then there are other variables like volume

Could you please show us one of these derivations?

 

[ZQrn]

Of course you can, the question is just to what accuracy you can.

I mean, the simplest formula I can give for the growth of an average stock is a linear formula that's just the growth of the economy. And if you take the limit to infinity you will find that if you wait long enough my formula does the expected.

I mean, the same thing happens when a rock falls and we use Newtonian mechanics. It's a limit, it's summing up the quantum indeterminacy of all those particles and using the expectancy value, in most cases it comes pretty damn accurate, but as well all know, theoretically the rock can just take the one in a gazillion chance and have all of its particles go upward and defy gravity. Stock market is no different, except that there are a lot less players in the stock market than particles in a rock.

Edit: THis thread shows in a funny way what the difference is between how physicists and mathematicians look at things.

You give a physicist some problem, he will try to solve it, or in this case correctly induce that a solution is not feasible at this juncture.

You give the same problem to a mathematician, he will say 'define predicting the stock market, what conditions must a solution meet to qualify as a prediction?'

 

[edward]

You folks really need to watch the Alchemists of Wall Street.



Many of the big investment banks are already doing what some here have questioned as being possible. It is all higher math.

I'll try to find a shorter version.

 

[ZQrn]

You folks really need to watch the Alchemists of Wall Street.



Many of the big investment banks are already doing what some here have questioned as being possible. It is all higher math.

I'll try to find a shorter version.

For some reason I actually trust what random people say in this thread more without having seen their credentials than that I trust the VPRO. You know these guys often obfuscate references to evolution right?

Also, the first guy already says an 'of course', which is hardly 'of course', a lot of people didn't make more money, they went bankrupt.

 

[edward]

For some reason I actually trust what random people say in this thread more without having seen their credentials than that I trust the VPRO. You know these guys often obfuscate references to evolution right?

Also, the first guy already says an 'of course', which is hardly 'of course', a lot of people didn't make more money, they went bankrupt.

You are not trusting the video because of the web site?? It was on other web sites, that was just the first one I found.

How about a shorter version from the WSJ, this one has a lot about the economy going in the tank. 3 minutes 41 seconds.

http://online.wsj.com/video/quants-...sis/53169DF3-31BD-42B4-AC13-445C50272120.html

 

[ZQrn]

You are not trusting the video because of the web site?? It was on other web sites, that was just the first one I found.

How about a shorter version from the WSJ, this one has a lot about the economy going in the tank. 4 minutes 41 seconds.

http://online.wsj.com/video/quants-...sis/53169DF3-31BD-42B4-AC13-445C50272120.html

No, because the VPRO produced it, and because it already starts with a pretty unbelievable claim it does not back up.

They're sensationalists, it's essentially the resident punker public broadcaster here.

Of course, all mainstream broadcasting distorts science like it's a party.

 

[Office_Shredder]

It is certainly true that applied mathematicians make models that predict what you should do to make money in the stock market. It's also true that these models have enough success that these guys get paid boatloads of money

 

[edward]

No, because the VPRO produced it, and because it already starts with a pretty unbelievable claim it does not back up.

They're sensationalists, it's essentially the resident punker public broadcaster here.

Of course, all mainstream broadcasting distorts science like it's a party.

OK I see your problem I didn't even know who or what VPRO was.

http://en.wikipedia.org/wiki/VPRO

On the other hand the documentary was right on the money...no pun intended. Watch the WSJ video, or are they not to be trusted either?

 

[Evo]

Perhaps this works when you are spreading millions across a ton of different stocks, but for the small investor that can only afford a few, I doubt that there would be any longterm success.

 

[ZQrn]

Perhaps this works when you are spreading millions across a ton of different stocks, but for the small investor that can only afford a few, I doubt that there would be any longterm success.

I guess the expectancy value of trusting the maths is higher than trusting the instincts of the average casual stock investor though.

That needn't imply that the expectancy value in either case is positive, of course.

 

[cronxeh]

I can tell you definitively that any math that you use for long term prediction of the stock market is going to be 100% wrong. There is absolutely, unequivocally, unfathomably, no doubt in that.


The only valuation you can use is the Warren Buffet method. Trust the people that run the company, trust the product they are selling, and have an outlook for the future. That is why he does not invest in tech stocks - considering India just came out with a $36 iPad, there is no clear certainty in the market. I even remember when I bought Apple for 78 bucks a share and offloaded it at 109. It was just a flimsy game, a short term fling.

Warren on the other hand, had a vision, with the exception of the airline industry

 

[ZQrn]

I can tell you definitively that any math that you use for long term prediction of the stock market is going to be 100% wrong. There is absolutely, unequivocally, unfathomably, no doubt in that.

Just as any long term prediction for the trajectory of a bullet, but people are still made to believe you can do that with calculus.

Also, define: 'long term', define 'wrong' et cetera et cetera.

 

[cronxeh]

Just as any long term prediction for the trajectory of a bullet, but people are still made to believe you can do that with calculus.

Also, define: 'long term', define 'wrong' et cetera et cetera.

Over 24 months, 'wrong' as in you will be off by anywhere between 2 and 3 standard deviations from your prediction. Oh and this does not apply to the really-really stable stocks like KO and MCD

 

[Office_Shredder]

That's strange, because long term the market grows, and that's pretty un-debatable (unless you're talking about long term to the point that we see the dollar collapse and the US is not a country or something) It's short term that's a problem