I'm trying to build a model in R that will let me predict when a costumer will purchase a product again. For example, the training data list customers who purchased bikes. I want to predict when these customers will purchase another bike in my store. However I don't know what kind of model can handle time and predict this kind of outcome. I thought about trying a buy till you die model, but this outputs the frequency of future purchases, and not when the next purchase will happen.
$\begingroup$
$\endgroup$
2
-
$\begingroup$ You could probably use a survival model. Instead of survival you are looking at the length of time until purchase. A Poisson model where you count the number of days until the event might also be appropriate here. $\endgroup$– StatsStudentCommented Jan 27, 2016 at 3:54
-
$\begingroup$ What does your data set look like? $\endgroup$– dsaxtonCommented Aug 28, 2016 at 23:09
Add a comment
|
1 Answer
$\begingroup$
$\endgroup$
@Michael Elma , i'm working on the same topic , i think you should collect all varaibles witch can affect customer purchase behaivour and after you doing a logistic regression (1/0). let me know more for your topic
