取兩列的平均值sql

Select Age, 

     (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Male') as avg_male, 
     (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Female') as avg_female 

from [dbo].[Transaction] t1 

group by Age ORDER BY Age

嗨，我有兩列avg_male和avg_female我如何取兩列的平均值並在第三列顯示。請幫助取兩列的平均值sql

來源

2014-02-17 vini

看看我的答案。要做第三欄，您可以簡單地進行相同的平均計算，而不考慮where子句中的性別。 – TheOneWhoPrograms

只需使用簡單的數學：

select avg_male 
,  avg_female 
,  (avg_male + avg_female)/2 
from transaction

編輯：

我認爲這應該工作。沒有SQLFiddle，但它應該做的事：

Select Age 
,  sum(case when ChildGender='Male' then ShoeSize else 0 end)/sum(case when ChildGender='Male' then 1 else 0 end) as avg_male, 
,  sum(case when ChildGender='Female' then ShoeSize else 0 end)/sum(case when ChildGender='Female' then 1 else 0 end) as avg_male, 
,  avg(ShoeSize) as avg_both 
from [dbo].[Transaction] t1 
group 
by  Age 
ORDER 
BY  Age

來源

2014-02-17 09:07:03

請您可以在同一個查詢中說明我寫的是我的真實表格是Transaction – vini

@vini：請參閱更新。 –

加權平均值是多少？例如，如果有10個男性和100個女性。 – MatBailie

使用(avg(x) + avg(y))/2，兩列將具有相同的重量，即使他們沒有條目相同數量。如果你想利用所有值的總算術平均在兩列，你需要他們的總數總結起來和將它們劃分：

SELECT (sum_male + sum_female)/(count_male + count_female) 
    FROM transaction

的SUM和COUNT聚合函數幫助你計算這些值：

(select sum(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Male') as sum_male, 
(select sum(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Female') as sum_female 
(select count(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Male') as count_male, 
(select count(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Female') as count_female

提供所有記錄任Male或FemaleChildGender， TheOneWhoPrograms's answer是一個更直接的解決方案。

來源

2014-02-17 09:09:23

感謝您參考我的答案。我已經修改了我的答案，以考慮他的桌子不僅僅是男性還是女性。 – TheOneWhoPrograms

爲什麼不使用avg函數？ – vini

@vini假設你已經有20個男孩，都是10歲，30個女孩，都是5歲，在你的數據庫中。 '（平均（男孩）+總和（女孩））/（總數（男孩）+總數（女孩））（平均（男孩）+平均（女孩））/ 2 =（10 + 5）/ 2 = 7.5' ）=（200 + 150）/（20 + 30）= 7'。與前者不同，後者的計算考慮到女孩多於男孩的事實，導致平均年齡較低。看到不同？沒有*正確*或*錯誤*，有兩種不同的公式，由您決定哪一個是您需要的。 –

以下應該工作。它和你做同樣的事情，但第三列沒有考慮到性別而做平均。

如果只有男性或女性兒童

Select Age, 

    (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and  t2.ChildGender='Male') as avg_male, 
    (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Female') as avg_female 
    (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age as avg_avg 
from [dbo].[Transaction] t1 

group by Age ORDER BY Age

如果有一些不男性或女性，你只需要男性的平均和女性

Select Age, 

    (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and  t2.ChildGender='Male') as avg_male, 
    (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age and t2.ChildGender='Female') as avg_female 
    (select avg(ShoeSize) from [dbo].[Transaction] t2 where t2.age = t1.Age AND (t2.ChildGender='Male' OR t2.ChildGender='Female') as avg_avg 
from [dbo].[Transaction] t1 

group by Age ORDER BY Age

來源

2014-02-17 09:11:08 TheOneWhoPrograms

;with CTE AS 
(
select 0 avg_male ,avg(ShoeSize) avg_female from [Transaction] ChildGender='Female' 
union 
select avg(ShoeSize) avg_male ,0 avg_female from [Transaction] ChildGender='Male' 
) 
select 
SUM(avg_male) avg_male,SUM(avg_female),(SUM(avg_male) avg_male+SUM(avg_female))/2 Total 
from CTE

來源

2014-02-17 09:18:22

一有用的一條知識是從集合中排除NULL。

這意味着AVG({1, 2, 3, NULL})是2. （1 + 2 + 3）/（3）。

以下使用CASE到NULLify某些記錄然後可以簡化事情。

SELECT 
    Age, 
    AVG(CASE WHEN ChildGender = 'Male' THEN ShoeSize ELSE NULL END) AS avgMale, 
    AVG(CASE WHEN ChildGender = 'Female' THEN ShoeSize ELSE NULL END) AS avgFemale, 
    AVG(          ShoeSize    ) AS avgAll 
FROM 
    [dbo].[Transaction] 
GROUP BY 
    Age

這招還與SUM()，COUNT()等

來源

2014-02-17 09:34:58 MatBailie

取兩列的平均值sql

回答

相關問題