SAS-and-HEC-Montreal-Cortex-Analytics-Simulation-Hackathon | A Simulation Game to Learn Predictive Analytics

Hackathon Ranking: Ranking

Hackathon Digital Badge: Badge

Project: Fundraising Campaign

HEC Montereal Canada Simulation Lab: ERPsimLab

Cortex Analytics Simulation: SAS Cortex

Venue: 1337 Coding Institute UM6P

Location: Ben Guerir, Morocco, North Africa

Organizers: Mohammed VI Polytechnic University, SAS, ERPsimLab HEC Montereal Canada

Date: December 2021

Time: 3 days 72 hours Hackathon

This project is a to create a predictive model to analyze the donor's behavior to maximize the operating surplus for the fundraising campaign. The model platform is created by SAS Analytics.

Project Scenario

You will be working on a fundraising campaign for a 12-year old, not-for-profit charitable organization (foundation) with a million members. The foundation has decided to add a direct contact campaign to its list of marketing activities. You will be using a predictive modeling software (SAS Enterprise Miner or SAS Visual Data Mining and Machine Learning) to predict how many and which individuals to target in the campaign. The objective is to fundraise the most in donation amount given the costs of calling members (sum of predicted amount given minus costs). You will be provided with the dataset of potential donors, and pre-built diagrams in the software, which will fit models based on previous behavior of donors (if they gave or not or how much they gave) and will also score donors to predict this year donation. The list of scored donors will be exported to an output file/report. Using this output, you will have to decide how many potential donors to target and will have to create a list of IDs of those potential donors. You will have to upload/submit the created list to the platform which will rank the submissions based on operating surplus – i.e., sum of donations minus the total cost of calling.

The detailed description is in pdf Hachathon materials folder uploded in the directory.

The predictive model we created uses decision tree and random forest machine learnin predictive agorithms to predict the chance of donation before and after contact. And we use multiple linear regression to predict the donation amount. The detailed model is in Cortex SAS.