Quantile Threshold Weekly (QtW)

YAML type: qtw | Category: Quantile

Compares weekly aggregates against a global weekly quantile.

Description

Same idea as Qt on weekly rolled-up values.

Quantile Threshold Weekly (QtW) illustration

Quantile Weekly first builds weekly aggregates, then compares the current week to a historical weekly percentile of those aggregates.

Choose aggregateMode to match how the business reads weekly KPIs. This keeps distribution-based alerting while smoothing day-to-day noise.

Choose aggregateMode and percentile.

Configure detectors on the data view Analysis tab or in YAML. Validate with Data Preview before enabling production schedules.

Parameter	Default	Description
`percentile`	`95`	Weekly historical quantile
`mode`	`above`	Direction -- see Allowed values
`aggregateMode`	`sum`	Period aggregation -- see Allowed values

Value	Meaning
`above`	Flag values above the quantile boundary
`below`	Flag values below the quantile boundary

both is not implemented for quantile detectors (qt, qtw, qtm).

Used when the detector compares period totals (week / month / quarter), not raw daily points.

Value	Meaning
`sum`	Compare sums of daily values in the period (default)
`average`	Compare averages of daily values in the period

Most detectors also accept optional series filters in params:

Parameter	Purpose
`maxGapAmount`	Skip series when gap amount is too high (`null` = no limit). Catalog create-default is often `5`.
`minValue`	Skip series whose average value is below this (`null` = no limit). Catalog create-default is often `1000`.
`maxValue`	Skip series whose average value is above this (optional; omit/`null` = no limit)
`useSpecialDates`	When `true`, suppress anomalies that overlap special dates

When a filter skips a series, Data Preview shows the reason.

analysis:
  detectors:
    - type: qtw
      params:
        percentile: 95
        mode: above
        aggregateMode: sum

Reduces daily noise while keeping distribution-based alerting.