SQL SERVER中PERCENTILE_CONT和PERCENTILE_DISC

SQL SERVER中PERCENTILE_CONT和PERCENTILE_DISC

PERCENTILE_CONT和PERCENTILE_DISC

看下面一組SQL語句：

WITH testas(    select N'LeeWhoeeUniversity' as name,10 as score    UNION ALL    select N'LeeWhoeeUniversity',20    UNION ALL    select N'LeeWhoeeUniversity',30    UNION ALL    select N'LeeWhoeeUniversity',40    UNION ALL    select N'LeeWhoeeUniversity',50    UNION ALL    select N'DePaul',60    UNION ALL    select N'DePaul',70    UNION ALL    select N'DePaul',80    UNION ALL    select N'DePaul',90    UNION ALL    select N'DePaul',100)select name,score,PERCENT_RANK() over(partition by name order by score) as per_rnk,PERCENTILE_CONT(0.5) WITHIN GROUP (ORDER BY score) over(partition by name) as percont0_5,PERCENTILE_CONT(0.6) WITHIN GROUP (ORDER BY score) over(partition by name) as percont0_6,PERCENTILE_CONT(0.7) WITHIN GROUP (ORDER BY score) over(partition by name) as percont0_7,PERCENTILE_CONT(0.75) WITHIN GROUP (ORDER BY score) over(partition by name) as percont0_75,PERCENTILE_DISC(0.5) WITHIN GROUP (ORDER BY score) over(partition by name) as perdist0_5,PERCENTILE_DISC(0.6) WITHIN GROUP (ORDER BY score) over(partition by name) as perdist0_6,PERCENTILE_DISC(0.7) WITHIN GROUP (ORDER BY score) over(partition by name) as perdist0_7,PERCENTILE_DISC(0.75) WITHIN GROUP (ORDER BY score) over(partition by name) as perdist0_75from test

運行結果：

name                          score    per_rnk    percont0_5    percont0_6    percont0_7    percont0_75    perdist0_5    perdist0_6    perdist0_7    perdist0_75
LeeWhoeeUniversity    10           0                 30                       34                     38                        40                     30                 30                  40                       40
LeeWhoeeUniversity    20           0.25           30                       34                      38                       40                      30                30                   40                       40
LeeWhoeeUniversity    30           0.5              30                       34                      38                       40                      30                30                   40                       40
LeeWhoeeUniversity    40           0.75            30                       34                      38                      40                        30               30                   40                       40
LeeWhoeeUniversity    50           1                 30                       34                      38                       40                       30                30                   40                       40
DePaul                    60            0                80                       84                      88                       90                       80                80                   90                       90
DePaul                    70            0.25          80                        84                      88                      90                       80                 80                  90                       90
DePaul                    80            0.5             80                       84                      88                      90                       80                 80                  90                        90
DePaul                    90            0.75          80                        84                      88                      90                       80                 80                  90                       90
DePaul                  100           1                80                        84                       88                     90                        80                80                  90                        90

簡單理解，PERCENT_RANK前面介紹過，求出score百分比後，PERCENTILE_CONT和PERCENTILE_DISC就是根據百分比求出對應的score。但不同的是，某一百分比沒有對應的score時，PERCENTILE_CONT會根據百分比的偏差計算出一個新值，此值可能並不存在於score中。PERCENTILE_DISC得出的是偏向於最近百分比對應的score值，因此此值肯定存在於score中。

如PERCENTILE_CONT(0.6)對應的值為34，偏差值我猜測是這樣計算出的：百分比75%和50%PERCENTILE_CONT計算出的值分別是40和30.那麼等式：(40-30)/(0.75-0.5)=新偏差值/0.6-0.5。PERCENTILE_CONT(0.6)在50%上的偏差值=4。所以，PERCENTILE_CONT(0.6)對應PERCENTILE_CONT(0.5)+4=34。

再看PERCENTILE_DISC(0.6)，直接將最近的0.5計算出的，PERCENTILE_DISC值取過來了。