Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: THEend8_
IEOR E4601: Group Project Warmup
This is a warmup to the main project to make sure everyone is comfortable with the logistics.
You are a seller with infinite inventory of a product and facing a logistic demand with unknown
parameter. Whenever a customer t arrives, you offer a price pt, and the customer accepts with
probability
e−βpt
1 + e−βpt
,
and rejects otherwise. You get a revenue of pt if customer t accepts. The parameter β is unknown.
Your goal is to design a policy that computes pt as a function of the historical observations up to
time t such that the total expected revenue is maximized.
You will need to write a function strategy() in python which outputs the price you want to
offer to the current customer. As we simulate your code, we will write the historical data in the file
historical data.csv in your home directory where each line t corresponds to user t and records
price pt and Xt ∈ {0, 1} to denote whether user t accepted pt (Xt = 1) or rejected (Xt = 0). The
function strategy can read the file historical data.csv to compute the price for next customer.
You will be able to observe the data and change your code for the function during the simulation.
We will query your function in the first half of every hour (i.e. 12-12:30, 1-1:30 and so on...), so
make sure you have a working code saved in the directory during that time. Note in this warmup
exercise, the teams are not playing against each other which will be the focus of the main project.
You must to submit a PDF that explains your approach on Gradescope in addition to the code on
the server gyanada.ieor.columbia.edu.