FA51 - Correlation and Regression

📊 This project explores how nutritional factors predict carbohydrate content in fast food items using correlation and regression analysis.
Developed as part of FA51: Formal Analysis at Minerva University.

---

🔑 Key Highlights

• Built a multivariate regression model to predict carbohydrate content
• Identified calories, fiber, and sugar as the strongest predictors
• Achieved an adjusted R² of 0.792 → model explains ~79% of variability

---

🚀 Quick Access

• 📂 Dataset
• 📑 Final Report (PDF)
• 💻 Jupyter Notebook on Google Colab

---

🛠️ Tools & Methods

• Python (pandas, numpy, matplotlib, seaborn, statsmodels)
• Data cleaning, correlation analysis, regression modeling
• Validation with p-values & confidence intervals

---

🌍 Impact

This project shows how data analysis can support healthier food decisions by predicting carbohydrate content in meals. It also demonstrates my ability to work with data science tools, build models, and communicate insights clearly.