Window

PERCENTILE_DISC

BigQuery

Mengembalikan nilai aktual pada percentile terdekat tanpa interpolasi.

Tipe hasil: Same as input expressionDiperbarui: 7 Jan 2026

Syntax

SQL

PERCENTILE_DISC(expression, percentile) OVER ([PARTITION BY partition_expression])

Parameter

expressionany orderablewajib

Kolom yang bisa diurutkan

percentileFLOAT64wajib

Nilai percentile antara 0 dan 1

PARTITION BYexpressionopsional

Membagi data ke dalam grup

Contoh Penggunaan

Median dengan Nilai Aktual

SQL

1 SELECT
2   department,
3   name,
4   salary,
5   PERCENTILE_DISC(salary, 0.5) OVER (PARTITION BY department) as median_salary
6 FROM `project.dataset.employees`;

Median yang merupakan nilai gaji aktual.

Hasil

department	name	salary	median_salary
Sales	David	50000	80000
Sales	Bob	80000	80000
Sales	Alice	120000	80000

Discrete vs Continuous

SQL

1 SELECT DISTINCT
2   category,
3   PERCENTILE_DISC(price, 0.5) OVER (PARTITION BY category) as median_disc,
4   PERCENTILE_CONT(price, 0.5) OVER (PARTITION BY category) as median_cont
5 FROM `project.dataset.products`;

Perbandingan DISC (nilai aktual) vs CONT (interpolasi).

Fungsi Terkait

STDDEVVARIANCECORRCOVAR_POPCOVAR_SAMPPERCENTILE_CONTSTDDEV_POPSTDDEV_SAMPVAR_POPVAR_SAMP

Kembali ke BigQuery Functions

1	SELECT
2	department,
3	name,
4	salary,
5	PERCENTILE_DISC(salary, 0.5) OVER (PARTITION BY department) as median_salary
6	FROM `project.dataset.employees`;

1	SELECT DISTINCT
2	category,
3	PERCENTILE_DISC(price, 0.5) OVER (PARTITION BY category) as median_disc,
4	PERCENTILE_CONT(price, 0.5) OVER (PARTITION BY category) as median_cont
5	FROM `project.dataset.products`;