0
$\begingroup$

I'm building a project that gathers the maximum amount of historic data about a certain company and try to predict its future market stock values.

I have company's 1 years stock values like open , close, low, high ,volume for each day.

Using these much data, can I predict the stock value?

What can be the approach for this? Is a simple NN with back-propagation training the best I can hope for? Stock data looks like -

https://www.google.com/finance/historical?q=NYSE%3ABME&ei=DSIeVtGMAYrvuATEsZfACA

$\endgroup$
  • $\begingroup$ This is a billion dollar question :) I suppose it depends on the particular series. Some are really tricky, but others are more stable. $\endgroup$ – Felipe Gerard Dec 3 '15 at 6:43
  • 5
    $\begingroup$ Of course you can predict the stock values. The relevant question is how accurate your predictions will be. $\endgroup$ – whuber Dec 3 '15 at 15:26
1
$\begingroup$

Yes, this is possible. You just need better models or better data than the remaining participants.

It is possible to show this using historical data and applying modern algorithms that were not available at that time. This is worked out in more detail in http://dx.doi.org/10.6084/m9.figshare.1287224 .

| cite | improve this answer | |
$\endgroup$
  • $\begingroup$ it is only for viewing auto corelation only or for prediction also? $\endgroup$ – user123 Dec 4 '15 at 6:59
  • $\begingroup$ Predictions based on auto correlations. $\endgroup$ – user36160 Dec 5 '15 at 18:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.