Inspiration

The idea behind our project was to assess the causal effect of promotions on car sales. These estimates are more meaningful than looking at correlations because confounders are being taken into account. While this effect can be estimated through simple regression models,

What it does

The application is a webapp where you would input different parameters to our backend and we would compute whether the parameters would be a causation of improvements in sales.

How we built it

We decided make a webapp with the web backend completed using python Flask and the machine learning portion done using scikit-learn. Given the 8 hours time constraints and overall unfamiliarity with the frontend, we decided to make a very simple interface using flask-bootstrap.

Challenges we ran into

The data that we wanted was not easily accessible; in other words, it was very difficult to parse the data from many databases in the system. For instance, it seems that none of the clients have had the opportunity to been exposed to promotional offers.

Accomplishments that we're proud of

What we learned

With very limited frontend experience, we learned some Javascript to perform some front-end backend requests and manage to interface correctly.

What's next for Promotion Analysis

Built With

Share this project:

Updates