Outliers are values that are very different from the rest of the data.
🔴 Why Outliers Matter:
They can:
- Skew averages
- Break models
- Mislead insights
🔍 Example:
Marks dataset:
- 70, 75, 80, 85, 90
- 500 (outlier)
Clearly unrealistic.
🔎 Methods to Detect Outliers:
1. Visual Methods
- Box plots
- Scatter plots
5
2. Statistical Methods
- Z-score
- IQR (Interquartile Range)
⚙️ What to Do with Outliers?
- Remove them (if clearly wrong)
- Cap them (limit extreme values)
- Keep them (if they are meaningful, e.g., high-income individuals)
💡 Key Idea:
Not all outliers are bad—some are valuable insights.