📊 TurnoverForecasting – AI Revenue Forecasting for SAP SE Link to heading

🔄 Overview Link to heading

This project builds an AI-powered turnover forecasting system for SAP SE, using a univariate SARIMA model. It demonstrates how reliable forecasts can be generated from minimal data — only historical revenue — making it ideal for early-stage AI adoption, SMEs, and strategic financial planning.

📂 Dataset Link to heading

🔗 Source: Top 12 German Companies Financial Data (Kaggle)
📅 Description: A financial dataset focused on German enterprises, with historical turnover values for SAP SE used to train and validate the forecasting model.

🗃️ Repository & Deployment Link to heading

🔗 GitHub Repository: View on GitHub
🚀 Live Demo: Try on Hugging Face

✨ Features Link to heading

• 🗕️ Accurate revenue forecasts up to 6 quarters ahead
• 🎯 Dynamic controls for forecast horizon and confidence intervals
• 🧠 Clean, Gradio-based interactive dashboard
• 📱 Mobile-friendly single-column layout
• 📈 Insightful visuals: Training, Validation, Test & Forecasts
• 🧩 Ideal for strategic planning, budgeting, and executive reporting

🛠️ Tools and Libraries Link to heading

• Language: Python
• Libraries: pandas, numpy, statsmodels, plotly
• Deployment: gradio, hosted on Hugging Face Spaces

🔧 How to Run Locally Link to heading

# Clone the repository
git clone https://github.com/Sharma-Pranav/Portfolio.git

# Navigate to the project directory
cd projects/TurnoverForecasting

# Install dependencies
pip install -r requirements.txt

# Run the Gradio app
python app.py

📊 Results Link to heading

• 📊 Reliable quarterly forecasts of SAP SE revenue
• ✅ Model validated using walk-forward validation
• 📉 Clear visualization of historical vs. forecasted revenue
• 💼 Actionable insights for financial strategy and planning