0
$\begingroup$

Let's assume I have recorded data from 1/1/2000 to 30/12/2018 it could be stock price dataset or rate (interest, currency or employment..etc)

And I want to predict price/rate for given date (1/1/2015 - 30/12/2018). I read a few papers and checked some examples and I noticed almost all of them just use 2 variables Date & target variable(the variable that we want to predict pric/rate) concerted to time series.

My question is, is this the only approach? What about the other variables won't they affect my prediction?

When it comes to these cases (dealing with time series), how can I use the other variables to predict rate/price?

$\endgroup$
1
  • $\begingroup$ A good source of idea would be kaggle competitions. There is currently one (JPX comp.) with some decent feature engineering approaches. (disclaimer: I wrote some of those notebooks). $\endgroup$ Jun 19, 2022 at 13:27

1 Answer 1

0
$\begingroup$

Most of these self-proclaimed data science tutorials on stock market price prediction are wrongly taught. I recently found this out using this video on youtube -> https://www.youtube.com/watch?v=xOcyV5Q2G5I. There are far more important variables that are needed to predict stock price than the previous price lol, for example - volume traded etc.

$\endgroup$
1
  • 1
    $\begingroup$ Your answer could be improved with additional supporting information. Please edit to add further details, such as citations or documentation, so that others can confirm that your answer is correct. You can find more information on how to write good answers in the help center. $\endgroup$
    – Community Bot
    Jun 16, 2022 at 19:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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