SQL Server Pivot on multiple fields - sql

Question

I have searched this website for all possible solutions but still can't find an answer for my Pivot problem.
I have a table with the following data.
Portfolio | Date | TotalLoans | ActiveLoans | TotalBalance
--------------------------------------------------------------------
P1 | 2015-12-31 | 1,000 | 900 | 100,000.00
P1 | 2015-11-30 | 1,100 | 800 | 100,100.00
P1 | 2015-10-31 | 1,200 | 700 | 100,200.00
I am trying to create a pivot with the following output (only where Portfolio = P1)
Field | 2015-12-31 | 2015-11-30 | 2015-10-31 |
-----------------------------------------------------
TotalLoans | 1,000 | 1,100 | 1,200 |
ActiveLoans | 900 | 800 | 700 |
TotalBalance | 100,000 | 100,100 | 100,200 |
Ideally, I am looking for a dynamic pivot, but a static query would do as well and I can try a dynamic query out of that.

Answer 1

You need first to UNPIVOT your table. You can do it using this query:
SELECT Portfolio, [Date], Val, ColType
FROM (SELECT Portfolio,
[Date],
TotalLoans,
ActiveLoans,
TotalBalance
FROM mytable
WHERE Portfolio = 'P1') AS srcUnpivot
UNPIVOT (
Val FOR ColType IN (TotalLoans, ActiveLoans, TotalBalance)) AS unpvt
Output:
Portfolio Date Val ColType
===============================================
P1 2015-12-31 1000 TotalLoans
P1 2015-12-31 900 ActiveLoans
P1 2015-12-31 100000 TotalBalance
P1 2015-11-30 1100 TotalLoans
P1 2015-11-30 800 ActiveLoans
P1 2015-11-30 100100 TotalBalance
P1 2015-10-31 1200 TotalLoans
P1 2015-10-31 700 ActiveLoans
P1 2015-10-31 100200 TotalBalance
Note: All unpivoted fields must be of the same type. The query above assumes a type of int for all fields. If this is not the case then you have to use CAST.
Using the above query you can apply PIVOT:
SELECT Portfolio, ColType, [2015-12-31], [2015-11-30], [2015-10-31]
FROM (
... above query here ...
PIVOT (
MAX(Val) FOR [Date] IN ([2015-12-31], [2015-11-30], [2015-10-31])) AS pvt

Answer 2

This is Giorgos Betsos solution as dynamic SQL. This will deal without the need to write the date values explicitly.
Please: If you like this: Do not mark this solution as accepted, set the acceptance to Giorgos Betsos. There's the hard work! But you may vote on it :-)
CREATE TABLE #tbl(Portfolio VARCHAR(10),[Date] DATE,TotalLoans DECIMAL(10,4),ActiveLoans DECIMAL(10,4),TotalBalance DECIMAL(10,4));
INSERT INTO #tbl VALUES
('P1','20151231',1000,900,100000.00)
,('P1','20151130',1100,800,100100.00)
,('P1','20151031',1200,700,100200.00);
DECLARE #pvtColumns VARCHAR(MAX)=
(
STUFF(
(
SELECT DISTINCT ',['+CONVERT(VARCHAR(10), [Date] ,126) + ']'
FROM #tbl
FOR XML PATH('')
)
,1,1,'')
);
DECLARE #cmd VARCHAR(MAX)=
'SELECT Portfolio, ColType, ' + #pvtColumns +
' FROM (
SELECT Portfolio, [Date], Val, ColType
FROM (SELECT Portfolio,
[Date],
TotalLoans,
CAST(ActiveLoans AS DECIMAL(10,4)) AS ActiveLoans,
TotalBalance
FROM #tbl AS mytable
WHERE Portfolio = ''P1'') AS srcUnpivot
UNPIVOT (
Val FOR ColType IN (TotalLoans, ActiveLoans, TotalBalance)) AS unpvt
) AS srcPivot
PIVOT (
MAX(Val) FOR [Date] IN (' + #pvtColumns + ')) AS pvt';
EXEC (#cmd);

Answer 3

You need to use Dyanmic SQL and construct them. See examples at http://social.technet.microsoft.com/wiki/contents/articles/17510.t-sql-dynamic-pivot-on-multiple-columns.aspx
Here is the code
CREATE procedure CrossTab
(
#select varchar(2000),
#PivotCol varchar(100),
#Summaries varchar(100),
#GroupBy varchar(100),
#OtherCols varchar(100) = Null
)
AS
set nocount on
set ansi_warnings off
declare #sql varchar(8000)
Select #sql = ''
Select #OtherCols= isNull(', ' + #OtherCols,'')
create table #pivot_columns (pivot_column_name varchar(100))
Select #sql='select ''' + replace( + #PivotCol,',',''' as pivot_column_name union all select ''')+''''
insert into #pivot_columns
exec(#sql)
select #sql=''
create table #pivot_columns_data (pivot_column_name varchar(100),pivot_column_data varchar(100))
Select #PivotCol=''
Select #PivotCol=min(pivot_column_name) from #pivot_columns
While #PivotCol>''
Begin
insert into #pivot_columns_data(pivot_column_name,pivot_column_data)
exec
(
'select distinct ''' + #PivotCol +''' as pivot_column_name, convert(varchar(100),' + #PivotCol + ') as pivot_column_data from
('+
#select
+'
) T'
)
Select #PivotCol=min(pivot_column_name) from #pivot_columns where pivot_column_name>#PivotCol
end
select
#sql = #sql + ', ' +
replace(
replace(
#Summaries,'(','(CASE WHEN ' + Pivot_Column_name + '=''' +
pivot_column_data + ''' THEN '
),
')[', ' END) as [' + pivot_column_data
)
from #pivot_columns_data
order by pivot_column_name
exec
(
'select ' + #GroupBy +#OtherCols +#sql +
' from (
'+
#select
+'
) T
GROUP BY ' + #GroupBy
)
drop table #pivot_columns
drop table #pivot_columns_data
set nocount off
set ansi_warnings on
Usage
EXEC CrossTab
'SELECT LastName, OrderDate,shipcountry FROM northwind..Employees Employees
INNER JOIN northwind..Orders Orders ON (Employees.EmployeeID=Orders.EmployeeID) ',
'shipcountry,Year(OrderDate)',
'Count(LastName)[]',
'LastName'

Answer 4

Do it in several steps:
if object_id('tempdb..#Data') is null
CREATE TABLE #Data
([Portfolio] varchar(2), [Date] datetime,
[TotalLoans] decimal(9,2), [ActiveLoans] int, [TotalBalance] decimal(9,2))
;
INSERT INTO #Data
([Portfolio], [Date], [TotalLoans], [ActiveLoans], [TotalBalance])
VALUES
('P1', '2015-12-31', 1000, 900, 100000.00),
('P1', '2015-11-30', 1100, 800, 100100.00),
('P2', '2015-10-31', 1200, 700, 100200.00)
;
WITH Transposed AS (
--First reorganise the data, creating unions like this, by column
--Assumption is that you are not interested in [Portfolio]
SELECT [Portfolio], [Date], [TotalLoans] AS Amount, 'TotalLoans' Field FROM #Data
UNION SELECT [Portfolio], [Date], [ActiveLoans], 'ActiveLoans' FROM #Data
UNION SELECT [Portfolio], [Date], [TotalBalance], 'TotalBalance' FROM #Data
)
SELECT Field, [2015-10-31], [2015-11-30], [2015-12-31] --You can build a string with all the dates from your original data source
FROM (
SELECT [Date], [Field], [Amount] FROM Transposed
) d
PIVOT (
MAX(Amount)
FOR [Date] IN ([2015-10-31], [2015-11-30], [2015-12-31])
) p