Oracle 分析函数详解（Analytic Functions）--示例部分
Analytic functions are commonly used in data warehousing environments. In the list of analytic functions that follows, functions followed by an asterisk (*) allow the full syntax, including the windowing_clause.
分析函数一般用于数据仓库环境。以下是分析函数列表，其中带星号的表示支持窗口语句windowing_clause.
AVG *CORR *
COVAR_POP *
COVAR_SAMP *
COUNT *
CUME_DIST
DENSE_RANK
FIRST
FIRST_VALUE *
LAG
LAST
LAST_VALUE *
LEAD
MAX *
MIN *
NTILE
PERCENT_RANK
PERCENTILE_CONT
PERCENTILE_DISC
RANK
RATIO_TO_REPORT
REGR_ (Linear Regression) Functions *
ROW_NUMBER
STDDEV *
STDDEV_POP *
STDDEV_SAMP *
SUM *
VAR_POP *
VAR_SAMP *
VARIANCE *
-------------------------------------------------------------------
1、AVG 为聚合函数用于求平均：
SELECT manager_id, last_name, hire_date, salary,
AVG(salary) OVER (PARTITION BY manager_id ORDER BY hire_date
ROWS BETWEEN 1 PRECEDING AND 1 FOLLOWING) AS c_mavg
FROM employees;
MANAGER_ID LAST_NAME HIRE_DATE SALARY C_MAVG
---------- ------------------------- --------- ---------- ----------
100 Kochhar 21-SEP-89 17000 17000
100 De Haan 13-JAN-93 17000 15000
100 Raphaely 07-DEC-94 11000 11966.6667
100 Kaufling 01-MAY-95 7900 10633.3333
100 Hartstein 17-FEB-96 13000 9633.33333
100 Weiss 18-JUL-96 8000 11666.6667
100 Russell 01-OCT-96 14000 11833.3333
2、CORR 返回一对表达式的相关系数：
SELECT employee_id, job_id,
TO_CHAR((SYSDATE - hire_date) YEAR TO MONTH ) "Yrs-Mns", salary,
CORR(SYSDATE-hire_date, salary)
OVER(PARTITION BY job_id) AS "Correlation"
FROM employees
WHERE department_id in (50, 80)
ORDER BY job_id, employee_id;
EMPLOYEE_ID JOB_ID Yrs-Mns SALARY Correlation
----------- ---------- ------- ---------- -----------
145 SA_MAN +08-07 14000 .912385598
146 SA_MAN +08-04 13500 .912385598
147 SA_MAN +08-02 12000 .912385598
148 SA_MAN +05-07 11000 .912385598
149 SA_MAN +05-03 10500 .912385598
150 SA_REP +08-03 10000 .80436755
151 SA_REP +08-02 9500 .80436755
152 SA_REP +07-09 9000 .80436755
153 SA_REP +07-01 8000 .80436755
154 SA_REP +06-05 7500 .80436755
155 SA_REP +05-06 7000 .80436755
3、COVAR_POP 返回一对表达式的总体协方差；
4、COVAR_SAMP 返回一对表达式的样本协方差；
5、COUNT 返回总行数：(每行对应的数据窗口是之前行幅度值不超过50，之后行幅度值不超过150)
SELECT last_name, salary,
COUNT(*) OVER (ORDER BY salary RANGE BETWEEN 50 PRECEDING
AND 150 FOLLOWING) AS mov_count FROM employees;
LAST_NAME SALARY MOV_COUNT
------------------------- ---------- ----------
Olson 2100 3
Markle 2200 2
Philtanker 2200 2
Landry 2400 8
Gee 2400 8
Colmenares 2500 10
Patel 2500 10
. . .
6、dense_rank 返回排名，用于TOPN查询：
查询假设薪资15500 、佣金5%的员工在employees表中排名
SELECT DENSE_RANK(15500, .05) WITHIN GROUP
(ORDER BY salary DESC, commission_pct) "Dense Rank"
FROM employees;
Dense Rank
-------------------
3
SELECT d.department_name, e.last_name, e.salary, DENSE_RANK()
OVER (PARTITION BY e.department_id ORDER BY e.salary) AS drank
FROM employees e, departments d
WHERE e.department_id = d.department_id
AND d.department_id IN ('30', '40');
DEPARTMENT_NAME LAST_NAME SALARY DRANK
----------------------- ------------------ ---------- ----------
Purchasing Colmenares 2500 1
Purchasing Himuro 2600 2
Purchasing Tobias 2800 3
Purchasing Baida 2900 4
Purchasing Khoo 3100 5
Purchasing Raphaely 11000 6
Human Resources Marvis 6500 1
7、first 当所查字段不是排序字段时返回分组范围内最大、最小值：
SELECT last_name, department_id, salary,
MIN(salary) KEEP (DENSE_RANK FIRST ORDER BY commission_pct)
OVER (PARTITION BY department_id) "Worst",
MAX(salary) KEEP (DENSE_RANK LAST ORDER BY commission_pct)
OVER (PARTITION BY department_id) "Best"
FROM employees
ORDER BY department_id, salary;
LAST_NAME DEPARTMENT_ID SALARY Worst Best
------------------- ------------- ---------- ---------- ----------
Whalen 10 4400 4400 4400
Fay 20 6000 6000 13000
Hartstein 20 13000 6000 13000
. . .
Gietz 110 8300 8300 12000
Higgins 110 12000 8300 12000
Grant 7000 7000 7000
SELECT last_name, department_id, salary,
MIN(salary) KEEP (DENSE_RANK FIRST ORDER BY commission_pct)
OVER (PARTITION BY department_id) "Worst",
MAX(salary) KEEP (DENSE_RANK LAST ORDER BY commission_pct)
OVER (PARTITION BY department_id) "Best"
FROM employees
ORDER BY department_id, salary;
8、fist_value 返回一组有序值中第一个值
SELECT department_id, last_name, salary, FIRST_VALUE(last_name)
OVER (ORDER BY salary ASC ROWS UNBOUNDED PRECEDING) AS lowest_sal
FROM (SELECT * FROM employees WHERE department_id = 90
ORDER BY employee_id);
DEPARTMENT_ID LAST_NAME SALARY LOWEST_SAL
------------- ------------- ---------- -------------------------
90 Kochhar 17000 Kochhar
90 De Haan 17000 Kochhar
90 King 24000 Kochhar
9、lag与lead函数是跟偏移量相关的两个分析函数，通过这两个函数我们可以取到当前行列的偏移N行列的值 lag可以看着是正的向上的偏移 lead可以认为负的向下的偏移
SELECT last_name, hire_date, salary,
LAG(salary, 1, 0) OVER (ORDER BY hire_date) AS prev_sal
FROM employees
WHERE job_id = 'PU_CLERK';
select deptno,
sal a,
lag(sal, 1, null) over(partition by deptno order by deptno) b
from scott.emp
SELECT last_name, hire_date,
LEAD(hire_date, 1) OVER (ORDER BY hire_date) AS "NextHired"
FROM employees WHERE department_id = 30;10、min/max 分别用于返回分组最小值/最大值：
SELECT manager_id, last_name, salary
FROM (SELECT manager_id, last_name, salary,
MAX(salary) OVER (PARTITION BY manager_id) AS rmax_sal
FROM employees) WHERE salary = rmax_sal;SELECT manager_id, last_name, hire_date, salary,
MIN(salary) OVER(PARTITION BY manager_id ORDER BY hire_date
RANGE UNBOUNDED PRECEDING) AS p_cmin
FROM employees;11、rank 类似于dense_rank 区别在于其排名数字不连续
SELECT RANK(15500) WITHIN GROUP (ORDER BY salary DESC) "Rank of 15500" FROM employees;
SELECT department_id, last_name, salary, commission_pct,
RANK() OVER (PARTITION BY department_id
ORDER BY salary DESC, commission_pct) "Rank"
FROM employees WHERE department_id = 80;12、row_number 和rownum差不多，功能更强一点（可以在各个分组内从1开始排序)
SELECT department_id, last_name, employee_id, ROW_NUMBER()
OVER (PARTITION BY department_id ORDER BY employee_id) AS emp_id
FROM employees;13、 RATIO_TO_REPORT 用来计算当前记录的指标expr占开窗函数over中包含记录的所有同一指标的百分比. 这里如果开窗函数的统计结果为null或者为0,就是说占用比率的被除数为0或者为null, 则得到的结果也为0
SELECT last_name, salary, RATIO_TO_REPORT(salary) OVER () AS rr
FROM employees
WHERE job_id = 'PU_CLERK';14、SUM
SELECT manager_id, last_name, salary,
SUM(salary) OVER (PARTITION BY manager_id ORDER BY salary
RANGE UNBOUNDED PRECEDING) l_csum
FROM employees;
to be continue...
------------------------
Dylan Presents.
------------------------------------
Dylan Presents.
