0
$\begingroup$

I have a general question regarding machine learning models. The idea is to predict what DATE the customer is likely to make transactions or purchases. Variables present in the data set are item, customer ids, prices, quote date, re-quote date, and transaction date. Y-variable would be transaction dates. Are there any models that predict dates when the customer likely to make transaction? So, in other words, can we make prediction of the dates based on quotes dates and/or re-quote dates. I know Time Series uses dates to make prediction for Sales,..etc (continuous target variables). I am not aware of any models which do this type of prediction.

$\endgroup$
3
  • 3
    $\begingroup$ If you re-state your problem to something like: "how many days it will take from $t_0$ time until purchase" the problem is simpler to solve - then you could use multiple methods starting from simple linear regression, to survival analysis etc. $\endgroup$
    – Tim
    Feb 4, 2015 at 20:10
  • 1
    $\begingroup$ Following re-statement by @Tim, you could then also consider autoregressive conditional duration (ACD) model. $\endgroup$ Feb 4, 2015 at 20:15
  • $\begingroup$ In particular, would $t_0$ be the quote date or re-quote date? $\endgroup$
    – Wayne
    Feb 4, 2015 at 20:43

0

Your Answer

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