How to find the second largest value in a field !

Hi,

i am taking the values from four tables ,

I am showing the Salesman in the descending order. according to their Sale Amount. by displaying in Descending, user can able to view the
salesman who sold for the highest amount.

Now I want to find the second highest Amount in the field.

for the highest and lowest we can use the Max and Min funtion.

for the second highest value, How can I write the query.

I already check the previous forums. But i couldn't get the idea.

Kindly reply me

Thank you very much,
Chock.Originally posted by chock
Hi,

i am taking the values from four tables ,

I am showing the Salesman in the descending order. according to their Sale Amount. by displaying in Descending, user can able to view the
salesman who sold for the highest amount.

Now I want to find the second highest Amount in the field.

for the highest and lowest we can use the Max and Min funtion.

for the second highest value, How can I write the query.

I already check the previous forums. But i couldn't get the idea.

Kindly reply me

Thank you very much,
Chock.
Well one way would be to say: what is the highest value after the highest value has been excluded (if you follow me):

SELECT MAX(amount)
FROM mytab
WHERE amount != (SELECT MAX(amount) FROM mytab);

Of course, you wouldn't want to use this recursive approach to get the 5th highest amount! For that, you could do:

SELECT amount FROM mytab m1
WHERE 4 =
(SELECT COUNT(DISTINCT amount) FROM mytab m2
WHERE m2.amount > m1.amount
);

i.e. get the amount for which there are exactly 4 higher amounts in the table.|||Hi,

You send me two queries, the first query I understand it. But in the second query

SELECT amount FROM mytab m1
WHERE 4 =
(SELECT COUNT(DISTINCT amount) FROM mytab m2
WHERE m2.amount > m1.amount
);
what's m1 and what's m2. In the previous query you didn't use the m1.

Actually Amount is the Field name we are going to compare and select.
and mytab is the Table Name.
I am new to this so I think i need some more o understand. can you please tell about the m1 and m2.

Thank you very much,
Chock.|||Originally posted by chock
Hi,

You send me two queries, the first query I understand it. But in the second query

SELECT amount FROM mytab m1
WHERE 4 =
(SELECT COUNT(DISTINCT amount) FROM mytab m2
WHERE m2.amount > m1.amount
);
what's m1 and what's m2. In the previous query you didn't use the m1.

Actually Amount is the Field name we are going to compare and select.
and mytab is the Table Name.
I am new to this so I think i need some more o understand. can you please tell about the m1 and m2.

Thank you very much,
Chock.
m1 and m2 are "aliases". I made them up, because I wanted to use the same table "mytab" twice in the same query and compare values. Without aliases the query would be:

SELECT amount FROM mytab
WHERE 4 =
(SELECT COUNT(DISTINCT amount) FROM mytab
WHERE mytab.amount > mytab.amount
);

... which will return no data, because the condition "WHERE mytab.amount > mytab.amount" is nonsense. What I want to say is "WHERE mytab.amount (in this subquery) > mytab.amount (in the main query)". Aliases allow you to do that.|||tony, you may have confused the issue by jumping from the second highest to the fifth

here's another way to get the row with the second highest value:select Salesman, SaleAmount
from SalesTable
where SaleAmount =
( select max(SaleAmount)
from SalesTable
where SaleAmount <
( select max(SaleAmount)
from SalesTable
)
)in english, "get the row where the SaleAmount is the highest SaleAmount that is less than the highest overall SaleAmount"

wouldn't want to nest that too deeply, eh

i believe a good optimiser will evaluate the innermost first (it is not correlated), then the next inner, then do a straight retrieval -- i could be wrong, though (it has happened, and optimizer performance is not my long suit)

rudy
http://r937.com

How to find a record

How to find a record in table without key? How to get specific record in table without key field and displaying on datagrid?

YOu explanation has to be a bit more specific. There is no problem in finfing any values (even if there is no "key" on the column) by just using a query for retrieving the data. I also do not understand what "How to get a specific record without a key filed" means, because if you want to have a specific record, you will have to provide an identifier for extracting / querying the data from the database.

HTH, Jens K. Suessmeyer.

http://www.sqlserver2005.de|||

hello,

you can search with the select statement, e.g.

NUMBER: SELECT * FROM STORES WHERE ZIPCODE = 21512

SELECT * FROM STORES WHERE ZIPCODE IN (12345, 23145, 23456)

SELECT * FROM STORES WHERE ZIPCODE BETWEEN 21431 AND 21531

TEXT: SELECT * FROM BOOKS WHERE DESCRIPTION = 'COOKING GUIDE'

SELECT * FROM BOOKS WHERE DESCRIPTION LIKE '%COOKING%'

DATE: SELECT * FROM SALES WHERE DATESOLD BETWEEN '2006-01-01' AND '2006-01-02'

SELECT * FROM SALES WHERE DATESOLD >= '2006-01-01' AND DATESOLD < '2006-01-02'

As you can see it depends on the type of data you are looking and on which tables you have to combine to gather the data

More info (tables involved, columns) and the report you are trying to make would help

(I believe this is a T-SQL topic though)