I have a complete Excel/CSV file that captures every client transaction for the full 2025 calendar year. My goal is to project 2026 performance from this dataset with a clear, month-by-month view and explicit bucket-flow percentages. What I’ll hand over • One well-structured Excel workbook (or CSV, if preferred) containing date-stamped transaction amounts, client IDs, and relevant classification fields. • A short glossary explaining each column so the model can be set up quickly. What I need back • A reproducible forecasting model (Python, R, or an advanced Excel solution—whichever suits you best) that ingests the 2025 file and outputs a full 2026 forecast. • Monthly breakdown of projected totals plus the bucket-flow % for each Bucket. • An exportable results sheet and a concise memo outlining the chosen method, key assumptions, and any caveats. Acceptance criteria 1. Forecast automatically recalculates if I paste a new 2025 data version into the same structure. 2. Mean Absolute Percentage Error (MAPE) from an in-sample back-test is disclosed so I can gauge reliability. 3. All formulas or code are unlocked and documented. Once delivered, I’ll validate the outputs against a small set of hand-computed checks; clear, transparent logic will let this go smoothly. If you’ve previously worked with time-series forecasting, ARIMA, Prophet, or similar tools, that background will be ideal, but I’m open to whichever technique fits the data best as long as it meets the above criteria and delivers an accurate monthly view with the requested bucket-flow percentages.