Sales Data Forecasting Analyst

I have a set of historical sales records—dates, units sold, revenue and a few categorical fields—that now need to drive reliable forecasts for the next two quarters. Your task is to clean the data, explore it for seasonality or anomalies, and then build a forecasting model that produces both point estimates and a visual projection of likely ranges.

You may work in Python (pandas, statsmodels, Prophet), R, or even Excel/Power Query if you can justify the approach; what matters is a clear, reproducible pipeline. Along the way I expect concise commentary on assumptions, feature engineering steps and the accuracy metrics you use to evaluate performance.

When you apply, attach a detailed project proposal: outline the methods you would test (e.g., ARIMA, exponential smoothing, machine-learning ensembles), a rough timeline, and any sample visual or metric you typically provide.

Deliverables:
• A cleaned version of the raw sales file
• The forecasting code or model, fully annotated
• A short report (PDF or slide deck) with charts, key findings and next-step recommendations

Acceptance criteria: the notebook or workbook must run end-to-end on my machine, forecasts should include confidence intervals, and the report must explain model choice and error scores in plain language.