0
$\begingroup$

I have data of bank branches and amount of revenue they have generated in a month. The data looks like this:

enter image description here

I am tasked to find the expected revenue for the branch for the next month using machine learning. Initially I was planning to use LSTM networks for such analysis, but I doubt its possible with such small amount of data.

I personally think machine learning is an overkill for such task. What would be the most appropriate way to predict the revenue for next month? I thought about increasing the amount of data by treating every branch as equal and using the row corresponding to each branch as separate instance for training (but I doubt that is a correct approach).

Any advice would be appreciated

Improve this question
$\endgroup$
2
  • $\begingroup$ I have also encountered similiar scenario but in a different domain, have you soloved the problem finally with the small amount of data? How did you solve it? Was it necessary to utilize machine learning model? Thank you. $\endgroup$
    – Anaconda
    Commented Apr 20, 2021 at 2:48
  • $\begingroup$ @Anaconda unfortunately with data this little only linear regression was possible, which ultimately is not the model that can successfully predict the next outcome $\endgroup$
    – Ach113
    Commented Apr 20, 2021 at 5:33

1 Answer 1

Reset to default
1
$\begingroup$

You might find the link helpful.

https://towardsdatascience.com/how-to-model-time-series-data-with-linear-regression-cd94d1d901c0

Improve this answer
$\endgroup$

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.