Use historical financial data to predict future revenue and expenses and generate forecast reports with visual projections for future growth.

• Python (pandas, statsmodels, matplotlib)
• Excel (for viewing and simple analysis of the CSV files)

File: data/financials_monthly.csv

Columns:

• Date – Month (from Jan 2020 to Dec 2024)
• Revenue – Monthly revenue
• Expenses – Monthly expenses

The data is synthetic but follows a realistic upward trend with seasonality and noise.

1. Data Preparation

   • Load the dataset and set Date as a time index.
   • Visualize historical revenue and expenses.

2. Model Selection

   • Use ARIMA (1,1,1) models separately for:
     • Revenue time series
     • Expenses time series

3. Forecasting

   • Generate forecasts for the next 12 months.
   • Save results to data/financials_forecast.csv.

4. Visualization

   • Plot:
     • Historical vs forecasted Revenue
     • Historical vs forecasted Expenses
   • Save plots to:
     • images/revenue_forecast.png
     • images/expenses_forecast.png

• data/financials_monthly.csv – Input historical data
• data/financials_forecast.csv – (Generated after running the script/notebook) 12‑month forecasts
• financial_forecasting.py – Python script to run the full pipeline
• financial_forecasting_notebook.ipynb – Jupyter Notebook with step‑by‑step analysis
• images/historical_revenue.png – Historical revenue plot
• images/historical_expenses.png – Historical expenses plot
• images/revenue_forecast.png – Forecast plot for revenue (generated when you run the script)
• images/expenses_forecast.png – Forecast plot for expenses (generated when you run the script)

1. Make sure you have Python 3 installed.
2. Install dependencies:

   pip install pandas matplotlib statsmodels

3. Navigate to this project folder (where financial_forecasting.py is located).
4. Run:

   python financial_forecasting.py

5. Check the outputs:
   • data/financials_forecast.csv
   • images/revenue_forecast.png
   • images/expenses_forecast.png

1. Install Jupyter:

   pip install notebook

2. Start Jupyter:

   jupyter notebook

3. Open financial_forecasting_notebook.ipynb and run all cells.

• Built a time series model (ARIMA) to forecast revenue and expenses.
• Produced 12‑month ahead forecasts for both metrics.
• Visualized growth trends and compared historical vs forecasted values.
• Demonstrated how forecasting can support budgeting and strategic planning.