Rules

Anomaly Formula: Interquartile-Range

What it does:
This formula detects outliers by checking if the latest value falls far outside the middle 50% of the data (between the 1st and 3rd quartiles). It uses a multiplier to control sensitivity.

When to use it:
Use this rule when you want to catch values that are significantly lower or higher than the rest of your data, without being too sensitive to small variations.

Examples

✅ Example 1: Outlier Detected (Increased)

• Anomaly Formula: InterquartileRange
• Change Type: Increased
• Multiplier: 1.5
• History: 25, 28, 29, 27, 30, 31, 29, 28, 29, 27, 100
• Explanation:
  • Q1 = 27
  • Q3 = 30
  • IQR = 3
  • Upper bound = 30 + (1.5 × 3) = 34.5
  • Latest value = 100 → Above upper bound → ✅ Alert triggered

✅ Example 2: Outlier Detected (Decreased)

• Anomaly Formula: InterquartileRange
• Change Type: Decreased
• Multiplier: 2
• History: 60, 62, 59, 61, 63, 58, 60, 59, 62, 5
• Explanation:
  • Q1 = 59
  • Q3 = 62
  • IQR = 3
  • Lower bound = 59 – (2 × 3) = 53
  • Latest value = 5 → Below lower bound → ✅ Alert triggered

❌ Example 3: Normal Case (Any)

• Anomaly Formula: InterquartileRange
• Change Type: Any
• Multiplier: 2
• History: 58, 84, 24, 96, 666, 14, 25, 23, 125, 39, 36, 48, 282, 10, 100, 25, 25, 28, 29, 28
• Explanation:
  • Q1 = 25
  • Q3 = 96.75
  • IQR = 71.75
  • Lower bound = -118.5
  • Upper bound = 240.25
  • Latest = 28 → ✅ Within bounds → ❌ No alert