WINDOW_MEDIAN: from Tableau to SQL

Calculates median over all regional marks, excluding null revenues. Below, 8 sample rows go through the calculation and through the SQL, and both return the same numbers.

Calculated field

WINDOW_MEDIAN(SUM([Revenue]))

Compute using: addressing Period; partitioning Region

Sample data

RegionPeriodRevenueCost
Harbor112070
Harbor2180100
Harbor3180120
Harbor4240140
Upland19050
Upland2150110
Upland3null90
Upland4330190

The same calculation as SQL

SELECT CAST((median("Revenue") OVER (PARTITION BY "Region" ORDER BY "Period" ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)) AS DOUBLE) AS result FROM input ORDER BY "Region", "Period"

Runs the example above.

Result

RegionPeriodRevenueCostresult
Harbor112070180
Harbor2180100180
Harbor3180120180
Harbor4240140180
Upland19050150
Upland2150110150
Upland3null90150
Upland4330190150

Same rows from the calculation and from the SQL in DuckDB.

Where a naive translation goes wrong

  • Omitting the region partition allows a calculation to consume marks from other regions.

    Naive SQL

    SELECT CAST((median("Revenue") OVER ( ORDER BY "Period" ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)) AS DOUBLE) AS result FROM input ORDER BY "Region", "Period"
    RegionPeriodRevenueCostresult
    Harbor112070180
    Harbor2180100180
    Harbor3180120180
    Harbor4240140180
    Upland19050180
    Upland2150110180
    Upland3null90180
    Upland4330190180

    Matches the calculation

    SELECT CAST((median("Revenue") OVER (PARTITION BY "Region" ORDER BY "Period" ROWS BETWEEN UNBOUNDED PRECEDING AND UNBOUNDED FOLLOWING)) AS DOUBLE) AS result FROM input ORDER BY "Region", "Period"
    RegionPeriodRevenueCostresult
    Harbor112070180
    Harbor2180100180
    Harbor3180120180
    Harbor4240140180
    Upland19050150
    Upland2150110150
    Upland3null90150
    Upland4330190150

Signature

WINDOW_MEDIAN(expression, [start, end])