原創(chuàng)于2009年08月02日，2009年10月22日遷移至此。

Oracle 分析函數(shù)——數(shù)據(jù)分布函數(shù)及 報(bào)表 函數(shù)

CUME_DIST

功能描述：計(jì)算一行在組中的相對(duì)位置， CUME_DIST 總是返回大于 0 、小于或等于 1 的數(shù)，該數(shù)表示該行在 N 行中的位置。例如，在一個(gè) 3 行的組中，返回的累計(jì)分布值為 1/3 、 2/3 、 3/3

SAMPLE ：下例中計(jì)算每個(gè)部門(mén)的員工按薪水排序依次累積出現(xiàn)的分布百分比

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

CUME_DIST() OVER (PARTITION BY department_id ORDER BY salary) AS cume_dist

FROM employees

NTILE

功能描述：將一個(gè)組分為 " 表達(dá)式 " 的散列表示，例如，如果表達(dá)式 =4 ，則給組中的每一行分配一個(gè)數(shù)（從 1 到 4 ），如果組中有 20 行，則給前 5 行分配 1 ，給下 5 行分配 2 等等。如果組的基數(shù)不能由表達(dá)式值平均分開(kāi)，則對(duì)這些行進(jìn)行分配時(shí)，組中就沒(méi)有任何 percentile 的行數(shù)比其它 percentile 的行數(shù)超過(guò)一行，最低的 percentile 是那些擁有額外行的 percentile 。例如，若表達(dá)式 =4 ，行數(shù) =21 ，則 percentile=1 的有 5 行， percentile=2 的有 5 行等等。

SAMPLE ：下例中把 6 行數(shù)據(jù)分為 4 份

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

NTILE(4) OVER (PARTITION BY department_id ORDER BY salary DESC) AS quartile

FROM employees

PERCENT_RANK

功能描述：和 CUME_DIST （累積分配）函數(shù)類(lèi)似，對(duì)于一個(gè)組中給定的行來(lái)說(shuō)，在計(jì)算那行的序號(hào)時(shí)，先減 1 ，然后除以 n-1 （ n 為組中所有的行數(shù)）。該函數(shù)總是返回 0 ～ 1 （包括 1 ）之間的數(shù)。

SAMPLE ：下例中如果 Khoo 的 salary 為 2900 ，則 pr 值為 0.6 ，因?yàn)? RANK 函數(shù)對(duì)于等值的返回序列值是一樣的

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

PERCENT_RANK() OVER (PARTITION BY department_id ORDER BY salary) AS pr

FROM employees

ORDER BY department_id,salary;

PERCENTILE_DISC

功能描述：返回一個(gè)與輸入的分布百分比值相對(duì)應(yīng)的數(shù)據(jù)值，分布百分比的計(jì)算方法見(jiàn)函數(shù) CUME_DIST ，如果沒(méi)有正好對(duì)應(yīng)的數(shù)據(jù)值，就取大于該分布值的下一個(gè)值。

注意：本函數(shù)與 PERCENTILE_CONT 的區(qū)別在找不到對(duì)應(yīng)的分布值時(shí)返回的替代值的計(jì)算方法不同

SAMPLE ：下例中 0.7 的分布值在部門(mén) 30 中沒(méi)有對(duì)應(yīng)的 Cume_Dist 值，所以就取下一個(gè)分布值 0.83333333 所對(duì)應(yīng)的 SALARY 來(lái)替代

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

PERCENTILE_DISC(0.7) WITHIN GROUP (ORDER BY salary ) OVER (PARTITION BY department_id) "Percentile_Disc",

CUME_DIST() OVER (PARTITION BY department_id ORDER BY salary) "Cume_Dist"

FROM employees<!-- [if gte vml 1]><v:shape id="_x0000_i1028" type="#_x0000_t75" style='width:400.5pt;height:270pt'><v:imagedata src="file:///C:/Users/bq_wang/AppData/Local/Temp/msohtmlclip1/01/clip_image007.png" o:title=""/></v:shape><![endif]--><!-- [if !vml]--><!-- [endif]-->

PERCENTILE_CONT

功能描述：返回一個(gè)與輸入的分布百分比值相對(duì)應(yīng)的數(shù)據(jù)值，分布百分比的計(jì)算方法見(jiàn)函數(shù) PERCENT_RANK ，如果沒(méi)有正好對(duì)應(yīng)的數(shù)據(jù)值，就通過(guò)下面算法來(lái)得到值：

RN = 1+ (P*(N-1)) 其中 P 是輸入的分布百分比值， N 是組內(nèi)的行數(shù)

CRN = CEIL(RN) FRN = FLOOR(RN)

if (CRN = FRN = RN) then

(value of expression from row at RN)

else

(CRN - RN) * (value of expression for row at FRN) +

(RN - FRN) * (value of expression for row at CRN)

注意：本函數(shù)與 PERCENTILE_DISC 的區(qū)別在找不到對(duì)應(yīng)的分布值時(shí)返回的替代值的計(jì)算方法不同

算法太復(fù)雜，看不懂了 L

SAMPLE ：在下例中，對(duì)于部門(mén) 60 的 Percentile_Cont 值計(jì)算如下：

P=0.7 N=5 RN =1+ (P*(N-1)=1+(0.7*(5-1))=3.8 CRN = CEIL(3.8)=4

FRN = FLOOR(3.8)=3

（ 4 - 3.8 ） * 4800 + (3.8 - 3) * 6000 = 5760

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

PERCENTILE_DISC(0.7) WITHIN GROUP (ORDER BY salary) OVER (PARTITION BY department_id) "Percentile_Disc",

PERCENTILE_CONT(0.7) WITHIN GROUP (ORDER BY salary) OVER (PARTITION BY department_id) "Percentile_Cont",

PERCENT_RANK() OVER (PARTITION BY department_id ORDER BY salary) "Percent_Rank"

FROM employees

總案例

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

CUME_DIST() OVER (PARTITION BY department_id ORDER BY salary) AS cume_dist, -- 數(shù)據(jù)分布百分比

NTILE(4) OVER (PARTITION BY department_id ORDER BY salary) AS quartile,-- 數(shù)據(jù)分布，以 NTILE 中的 exp 來(lái)計(jì)算

PERCENT_RANK() OVER (PARTITION BY department_id ORDER BY salary) AS pr,-- 數(shù)據(jù)分布百分比，從 0 開(kāi)始計(jì)

PERCENTILE_DISC(0.7) WITHIN GROUP (ORDER BY salary ) OVER (PARTITION BY department_id) "Percentile_Disc",-- 輸入的分布百分比值相對(duì)應(yīng)的數(shù)據(jù)值

PERCENTILE_CONT(0.7) WITHIN GROUP (ORDER BY salary) OVER (PARTITION BY department_id) "Percentile_Cont"-- 表達(dá)式太復(fù)雜了， ...

FROM employees

RATIO_TO_REPORT

功能描述：該函數(shù)計(jì)算 expression/(sum(expression)) 的值，它給出相對(duì)于總數(shù)的百分比，即當(dāng)前行對(duì) sum(expression) 的貢獻(xiàn)。

SAMPLE ：下例計(jì)算每個(gè)員工的工資占該類(lèi)員工總工資的百分比

SELECT

department_id,

first_name||' '||last_name employee_name,

salary,

RATIO_TO_REPORT(salary) OVER () AS rr

FROM employees

WHERE job_id = 'PU_CLERK';

REGR_ (Linear Regression) Functions

功能描述：這些線性回歸函數(shù)適合最小二乘法回歸線，有 9 個(gè)不同的回歸函數(shù)可使用。

REGR_SLOPE ：返回斜率，等于 COVAR_POP(expr1, expr2) / VAR_POP(expr2)

REGR_INTERCEPT ：返回回歸線的 y 截距，等于

AVG(expr1) - REGR_SLOPE(expr1, expr2) * AVG(expr2)

REGR_COUNT ：返回用于填充回歸線的非空數(shù)字對(duì)的數(shù)目

REGR_R2 ：返回回歸線的決定系數(shù)，計(jì)算式為：

If VAR_POP(expr2) = 0 then return NULL

If VAR_POP(expr1) = 0 and VAR_POP(expr2) != 0 then return 1

If VAR_POP(expr1) > 0 and VAR_POP(expr2 != 0 then

return POWER(CORR(expr1,expr),2)

REGR_AVGX ：計(jì)算回歸線的自變量 (expr2) 的平均值，去掉了空對(duì) (expr1, expr2) 后，等于 AVG(expr2)

REGR_AVGY ：計(jì)算回歸線的應(yīng)變量 (expr1) 的平均值，去掉了空對(duì) (expr1, expr2) 后，等于 AVG(expr1)

REGR_SXX ： 返回值等于 REGR_COUNT(expr1, expr2) * VAR_POP(expr2)

REGR_SYY ： 返回值等于 REGR_COUNT(expr1, expr2) * VAR_POP(expr1)

REGR_SXY: 返回值等于 REGR_COUNT(expr1, expr2) * COVAR_POP(expr1, expr2)

（下面的例子都是在 SH 用戶下完成的）

SAMPLE 1 ：下例計(jì)算 1998 年最后三個(gè)星期中兩種產(chǎn)品（ 260 和 270 ）在周末的銷(xiāo)售量中已開(kāi)發(fā)票數(shù)量和總數(shù)量的累積斜率和回歸線的截距

SELECT t.fiscal_month_number "Month", t.day_number_in_month "Day",

REGR_SLOPE(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_month_desc, t.day_number_in_month) AS CUM_SLOPE,

REGR_INTERCEPT(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_month_desc, t.day_number_in_month) AS CUM_ICPT

FROM sales s, times t

WHERE s.time_id = t.time_id

AND s.prod_id IN (270, 260)

AND t.fiscal_year=1998

AND t.fiscal_week_number IN (50, 51, 52)

AND t.day_number_in_week IN (6,7)

ORDER BY t.fiscal_month_desc, t.day_number_in_month;

SAMPLE 2 ：下例計(jì)算 1998 年 4 月每天的累積交易數(shù)量

SELECT UNIQUE t.day_number_in_month,

REGR_COUNT(s.amount_sold, s.quantity_sold)

OVER (PARTITION BY t.fiscal_month_number ORDER BY t.day_number_in_month)

"Regr_Count"

FROM sales s, times t

WHERE s.time_id = t.time_id

AND t.fiscal_year = 1998 AND t.fiscal_month_number = 4;

SAMPLE 3 ：下例計(jì)算 1998 年每月銷(xiāo)售量中已開(kāi)發(fā)票數(shù)量和總數(shù)量的累積回歸線決定系數(shù)

SELECT t.fiscal_month_number,

REGR_R2(SUM(s.amount_sold), SUM(s.quantity_sold))

OVER (ORDER BY t.fiscal_month_number) "Regr_R2"

FROM sales s, times t

WHERE s.time_id = t.time_id

AND t.fiscal_year = 1998

GROUP BY t.fiscal_month_number

ORDER BY t.fiscal_month_number;

SAMPLE 4 ：下例計(jì)算 1998 年 12 月最后兩周產(chǎn)品 260 的銷(xiāo)售量中已開(kāi)發(fā)票數(shù)量和總數(shù)量的累積平均值

SELECT t.day_number_in_month,

REGR_AVGY(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_month_desc, t.day_number_in_month)

"Regr_AvgY",

REGR_AVGX(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_month_desc, t.day_number_in_month)

"Regr_AvgX"

FROM sales s, times t

WHERE s.time_id = t.time_id

AND s.prod_id = 260

AND t.fiscal_month_desc = '1998-12'

AND t.fiscal_week_number IN (51, 52)

ORDER BY t.day_number_in_month;

SAMPLE 5 ：下例計(jì)算產(chǎn)品 260 和 270 在 1998 年 2 月周末銷(xiāo)售量中已開(kāi)發(fā)票數(shù)量和總數(shù)量的累積 REGR_SXY, REGR_SXX, and REGR_SYY 統(tǒng)計(jì)值

SELECT t.day_number_in_month,

REGR_SXY(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_year, t.fiscal_month_desc) "Regr_sxy",

REGR_SYY(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_year, t.fiscal_month_desc) "Regr_syy",

REGR_SXX(s.amount_sold, s.quantity_sold)

OVER (ORDER BY t.fiscal_year, t.fiscal_month_desc) "Regr_sxx"

FROM sales s, times t

WHERE s.time_id = t.time_id

AND prod_id IN (270, 260)

AND t.fiscal_month_desc = '1998-02'

AND t.day_number_in_week IN (6,7)

ORDER BY t.day_number_in_month;

Oracle分析函數(shù)六——數(shù)據(jù)分布函數(shù)及報(bào)表函數(shù)