QUANTILE
Syntax
Standard
Extended
QUANTILE( value, quant )
QUANTILE( value, quant
[ FIXED ... | INCLUDE ... | EXCLUDE ... ]
[ BEFORE FILTER BY ... ]
)
More info:
Description
Returns the precise quant
-level quantile (quant
should be in range from 0 to 1).
Argument types:
value
—Date | Datetime | Fractional number | Integer
quant
—Fractional number | Integer
Return type: Same type as (value
)
Example
Source data
Date | Temperature |
---|---|
2019-03-01 |
5.00 |
2019-03-02 |
0.00 |
2019-03-03 |
-3.00 |
2019-03-04 |
1.00 |
2019-03-05 |
4.00 |
2019-03-06 |
5.00 |
2019-03-07 |
1.00 |
2019-03-08 |
-6.00 |
2019-03-09 |
-6.00 |
2019-03-10 |
-4.00 |
2019-03-11 |
-2.00 |
2019-03-12 |
0.00 |
Formulas:
- 0.25 quantile:
QUANTILE([Temperature], 0.25)
; - 0.5 quantile:
QUANTILE([Temperature], 0.5)
; - 0.75 quantile:
QUANTILE([Temperature], 0.75)
.
Result
0.25 quantile | 0.5 quantile | 0.75 quantile |
---|---|---|
-3.00 |
0.00 |
4.00 |
Data source support
ClickHouse 21.8
, Oracle Database 12c (12.1)
, PostgreSQL 9.4
, YDB
.