Program Trading Stocks

Anyone know of a good book to read on building automated algorithms for trading stocks?  I don't have the time or patience to be sitting at the comp all day watching those damn graphs.

if(ishigh(price))
  sell(stock);
else if(islow(price))
  buy(stock);

SICP

PAIP

>>1
I don't have the time or patience to be sitting at the comp all day watching those damn graphs.
How about you just implement whatever calculations your pea brain does when looking at those stocks, since apart from boring you, it apparently does work? (Not likely, but you're a fag anyway)

Use instant.exe, it ships with OVER 200 ENVIRONMENT VARIABLES

http://www.instantexe.com/

>>2

And how do you implement ishigh() and islow()
?

>>7

int
ishigh(void)
{
  if(!islow()) {
    return 1;
  } else {
    return 0;
  }
}

int
islow(void)
{
  if(!ishigh()) {
    return 1;
  } else {
    return 0;
  }
}
}

>>8
I was drunk as shit when I wrote that. Just ignore the "Doesn't understand the Gregorian calendar." part. The rest is accurate.


int
islow(void)
{
(Post truncated.)

(Pos
(Post truncated.)

>>8

Infinite loop

>>12
Until your stack crashes that is.

>>13
stacks don't crash
this one will be optimized to an infinite loop

>>14
If you put islow and ishigh into different files, no, it won't be optimized

i am >>8 and i came here to say that YHBT. YHL. HAND.

>>16
No. You are wrong, because I am right.

Programming Collective Intelligence by Toby Segaran.

There are multiple ways to predict market behavior (regression, neural nets, decision trees, clustering, & many more), this book has a ton of algorithms. Most of the code examples are in Python which I know isn't too popular around here.

Toby is a girls name

Kunta Kinte

First read books like "A mathematician plays the stockmarket" then (if you can ignore the overbearing egoism) try "the black swan" (or "black swan events") which talk about how most such algorithms are, in fact, worthless bollocks.

Then lurk in Usenet alt.binaries.e-books.technical (or something close to that) because they often flood trading books, and programming books, and sometimes books about programming trading.

protip: if you could win at the stock markets with an algorithm, it would have been done by now

>>22
protip: it has been and is being done

>>23

And thousands of traders are being replaced by machines

>>24
And we're all millionaires.

>>24
No John, you are the machines.

Do you work for Goldman Sachs?  Can you give me details on their algorithms?

>>27
I work for Goldman Sachs and would be delighted to share my algorithms with you.

My algorithms mainly fall into two categories:

1) Numerical - Used to shape a bunch of data into something more usable. For example, taking a bunch of worldwide indices and putting them into a matrix, doing operations on them and then getting some data from the resulting matrix.

2) Machine Learning - Using data above to figure out trends in the stocks.

If you have any more specific questions, don't be afraid to reply and ask.

>>28
Could you please elaborate? Thanks.

All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy
 All work and no play makes Jack a dull boy