sas_graduate_projects

This repository contains SAS-based projects developed as part of my graduate coursework in Data Analytics. The scripts demonstrate applications of logistic regression, time series forecasting, and classification techniques using real-world datasets. Each project showcases statistical modeling, data preparation, and model evaluation performed in the SAS environment.

Projects

Super_Bowl.sas

Forecasts AFC and NFC championship winners using logistic regression. Includes stepwise variable selection, AIC model comparison, and Firth bias correction to improve convergence with small or imbalanced samples.

Requirements

• SAS 9.4+ or access to SAS Studio
• Compatible file formats: .csv, .xlsx
• No external libraries or macros required

Tools & Features Used

• SAS 9.4+ / SAS Studio – All code is written in Base SAS using standard procedures and data steps.
• PROC LOGISTIC – For binary classification with logistic regression and Firth bias correction.
• PROC EXPORT / IMPORT – For reading .csv and .xlsx datasets.
• Macro Variables & Stepwise Selection – To handle iterative modeling and variable filtering.
• Model Evaluation – Includes AIC, convergence checking, and prediction accuracy evaluation.
• No external packages or add-ons are required.

Getting Started

To run this project:

1. Clone or download this repository.
2. Open the .sas file in SAS Studio or your local SAS environment.
3. Update any file paths in libname or infile statements to match your local directory.
4. Run the script section-by-section to review model diagnostics and output.

Notes

• All projects are for educational and portfolio purposes.
• Results and models reflect exploratory work rather than production-level pipelines.

Other Resources

My Github R Projects

Author

Rebecca Calhoun

M.S. in Data Analytics (July 2025)
LinkedIn |