I am working with daily binary time series forecast as follows:

  • The target : purchase decision (0: not purchase, 1 purchase
  • Features: day, weekday, promotion, holiday,....

The objective is trying to forecast that the day have purchase or not! So, What algorithms can be used to address this problem! I also research about markov chain , Survival analysis.. but i am not sure it can be applied?

  • $\begingroup$ We will need details on: size of dataset, what features you have, etc $\endgroup$
    – GooJ
    Commented Sep 8, 2022 at 18:23
  • $\begingroup$ What makes this a time series? Is it the same person making the purchase decision each time? Then that is a time series. Is it a different person making the purchase decision each time? That is not a time series. $\endgroup$
    – Tripartio
    Commented Nov 19, 2022 at 19:31

1 Answer 1


Although the nature of your data might present a time-series format, it looks like you can frame it as a usual classifier, building a tabular dataset based on the features you point out, something like:

day weekday promotion holiday purchase_target
1 Monday 0.5 True 0
2 Tuesday 0.3 False 0
3 Wednesday 0 False 1

You can begin with simple binary classifiers, like naive bayes classifier, logistic regressor... and aslso have a look at a comparisson of some others in this link.

  • $\begingroup$ Thanks you! I also conduct this method. But the accuracy is not good! Any approach please suggest me! $\endgroup$
    – Sherry
    Commented Sep 5, 2021 at 7:46
  • $\begingroup$ this method does not guarantee accuracy by itself; you could add some feature engineering trying to add new explainable features, but first, make sure your dataset is realiable and you build a baseline model to have a reference $\endgroup$
    – German C M
    Commented Sep 5, 2021 at 9:55

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.