I have a numerical dataset that needs a careful scrub before any sense can be made of it. The main headache right now is outliers—they are skewing the story the numbers are trying to tell. I need you to detect, diagnose, and treat those extreme values using an approach you can justify statistically (Python / Pandas, R, or even advanced Excel are all fine as long as the method is transparent and reproducible). Once the data are healthy, I want straightforward descriptive statistics—think clear measures of central tendency, dispersion, and a concise written interpretation that highlights anything interesting the cleaned data reveals. No forecasting or trend-spotting models this time; just an honest summary of what the numbers say after the noise is removed. Deliverables: • Cleaned dataset with outliers handled (flagged or adjusted—explain your choice) • A short report or notebook showing the code/workflow plus the descriptive stats and narrative explanation • Quick hand-off guide so I can replicate the process on future datasets If this sounds like your kind of tidy-up, let’s get started.