0
$\begingroup$

I'm getting confused as how to proceed for this problem:

I need to predict a city's food supply, for number of orders, for the upcoming weeks based on previous weeks' data. What should I do? Do I need to use a time related algorithm or a simple algorithm?

$\endgroup$
2
  • 4
    $\begingroup$ A bit more context would be good. In principle this sounds like a time series problem. How long is your time series, what kind of variables/features do you have. How large is the variation over time? $\endgroup$
    – Peter
    Commented Jun 24, 2019 at 17:29
  • $\begingroup$ Yes..it is a time series problem..I have data based on no of weeks. $\endgroup$
    – Anu
    Commented Jul 8, 2019 at 15:52

1 Answer 1

0
$\begingroup$

If you treat this problem as a time-series one, you will gain more information for predictions, because it includes several time related components, and you can customize your orders predictions based on week day(week-end, Black Fridays)

If you treat this problem as linear regression, you will only be able to predict for the one upcoming week and not several weeks in advance, as they will only depend on the latest record you gave as the predictor.

analytics vidhya has great tutorials on time-series problems.

$\endgroup$
1
  • 1
    $\begingroup$ Thank you for the explanation, I'll work on it $\endgroup$
    – Anu
    Commented Jun 29, 2019 at 9:32

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.