I have a problem I'm trying to solve and was hoping someone smarter and more experienced may be able to help me out. Let me describe the problem. Thanks for any help in advance.
My company provides a service (let's call it A/C repair) in which quotes are generated on a regular basis. The pricing for the "service charge" on these quotes is dynamic can vary from customer to customer, office to office, and even quote to quote. This price can even be set by a salesperson with no knowledge of what has been done in the past which is where my problem arises. I need to provide this pricing recommendation.
I am trying to design a model that allows a user to see recommended pricing for a service based on historical win/loss trends over a specified time frame. So, as a salesperson, if I am quoting a service to a customer, I would like to select that service type of service and a time frame (past 30 days) and see a price recommendation. If I have been losing quotes to competitors in that time frame, that should have a negative impact on the suggestion (maybe we should lower price). If I am winning, that has a positive impact (let's stay the same or possibly increase price).
Based on my knowledge, this seems to be a regression problem, but the win/loss percentages and time frames are throwing me off. Can someone please help?