I have a dataset consisting of customer data of an e-commerce website and deals that have been shown to them for each item, the price of each item, the items they clicked on and the items they bought:(This data is from one session)
customer_id item_id deal items_clicked items_bought price($)
1 2 50%Off 1 0 2
1 4 No-deal 0 0 5
1 7 free-ship 0 0 1
1 3 20%Off 1 1 7
2 5 30%Off 1 1 6
2 7 No-deal 1 0 2
2 2 free-ship 0 0 4
2 1 free-ship 1 1 1
I want to know are and which deals are having a positive effect on customers, and find out what are the deals that customers who buy items prefer.
My initial thought is to run a logistic regression with target variable items_bought and dependent variable as.facotr(deal).
In this case, I don't think customer aggregation is needed, and I can use data as is. Is this the right path? Then does this also give me the preferred deals by buying customers?