PERCENTILE Function Examples | Teradata Vantage - 17.00 - Examples - Teradata Database

Teradata Vantageā„¢ - Time Series Tables and Operations

prodname
Advanced SQL Engine
Teradata Database
vrm_release
17.00
created_date
June 2020
category
Programming Reference
featnum
B035-1208-170K

Example: Using PERCENTILE

/*PTI Table*/
SELECT $TD_TIMECODE_RANGE, $TD_GROUP_BY_TIME, BUOYID, PERCENTILE(TEMPERATURE, 25), COUNT(*)
FROM OCEAN_BUOYS
WHERE TD_TIMECODE BETWEEN TIMESTAMP '2014-01-06 08:00:00' AND TIMESTAMP '2014-01-06 10:30:00'
GROUP BY TIME (MINUTES(10) AND BUOYID)
ORDER BY 3,2,1;

/*Non-PTI Table*/
SELECT $TD_TIMECODE_RANGE, $TD_GROUP_BY_TIME, BUOYID, PERCENTILE(TEMPERATURE, 25), COUNT(*)
FROM OCEAN_BUOYS_NONPTI
WHERE TIMECODE BETWEEN TIMESTAMP '2014-01-06 08:00:00' AND TIMESTAMP '2014-01-06 10:30:00'
GROUP BY TIME (MINUTES(10) AND BUOYID)
USING TIMECODE(TIMECODE)

The results are the same for both tables:

TIMECODE_RANGE GROUP BY TIME (MINUTES(10)) BUOYID Percentile (TEMPERATURE) Count(*)
('2014-01-06 08:00:00.000000+00:00', '2014-01-06 08:10:00.000000+00:00') 1 0 10 3
('2014-01-06 08:10:00.000000+00:00', '2014-01-06 08:20:00.000000+00:00') 2 0 10 2
('2014-01-06 09:00:00.000000+00:00', '2014-01-06 09:10:00.000000+00:00') 7 1 72 6
('2014-01-06 10:00:00.000000+00:00', '2014-01-06 10:10:00.000000+00:00') 13 44 43 10
('2014-01-06 10:10:00.000000+00:00', '2014-01-06 10:20:00.000000+00:00') 14 44 43 1

Example: Using the LOW Interpolation Scheme

/*PTI Table*/
SELECT $TD_TIMECODE_RANGE, $TD_GROUP_BY_TIME, BUOYID, PERCENTILE(TEMPERATURE, 25, LOW), COUNT(*)
FROM OCEAN_BUOYS
WHERE TD_TIMECODE BETWEEN TIMESTAMP '2014-01-06 08:00:00' AND TIMESTAMP '2014-01-06 10:30:00'
GROUP BY TIME (MINUTES(10) AND BUOYID)
ORDER BY 3,2,1;

/*Non-PTI Table*/
SELECT $TD_TIMECODE_RANGE, $TD_GROUP_BY_TIME, BUOYID, PERCENTILE(TEMPERATURE, 25, LOW), COUNT(*)
FROM OCEAN_BUOYS_NONPTI
WHERE TIMECODE BETWEEN TIMESTAMP '2014-01-06 08:00:00' AND TIMESTAMP '2014-01-06 10:30:00'
GROUP BY TIME (MINUTES(10) AND BUOYID)
USING TIMECODE(TIMECODE)
ORDER BY 3,2,1;

The results are the same for both tables:

TIMECODE_RANGE GROUP BY TIME (MINUTES(10)) BUOYID Percentile (TEMPERATURE) Count(*)
('2014-01-06 08:00:00.000000+00:00', '2014-01-06 08:10:00.000000+00:00') 1 0 10 3
('2014-01-06 08:10:00.000000+00:00', '2014-01-06 08:20:00.000000+00:00') 2 0 10 2
('2014-01-06 09:00:00.000000+00:00', '2014-01-06 09:10:00.000000+00:00') 7 1 72 6
('2014-01-06 10:00:00.000000+00:00', '2014-01-06 10:10:00.000000+00:00') 13 44 44 43
('2014-01-06 10:10:00.000000+00:00', '2014-01-06 10:20:00.000000+00:00') 14 44 43 1