Or maybe it would be best to just skip all of that insufferable reading and do it like this.

WITH SampleData (ID, Amount, Months) AS

(

SELECT 1, 143.23, 2 UNION ALL SELECT 2, 143.25, 6

),

Tally (n) AS

(

SELECT ROW_NUMBER() OVER (ORDER BY (SELECT NULL))

FROM (VALUES(0),(0),(0),(0),(0),(0),(0),(0),(0),(0)) a(n)

)

SELECT ID, Amount, Months, n, DividedAmt, TotalAmt

,CorrectedAmt=DividedAmt +

CASE WHEN ABS((TotalAmt-Amount)*100) <= n-1

THEN 0

ELSE .01*SIGN((TotalAmt-Amount))

END

FROM

(

SELECT ID, Amount, Months, n, DividedAmt

,TotalAmt=SUM(DividedAmt) OVER (PARTITION BY ID)

FROM

(

SELECT ID, Amount, Months, n

,DividedAmt=CAST(Amount/Months AS DECIMAL(10,2))

FROM SampleData a

CROSS APPLY

(

SELECT TOP (Months) n

FROM Tally

ORDER BY n

) b

) a

) a;

Note that most of the work (in fact both of the CTEs and the CROSS APPLY of the inner query, is just setting up the data/exploding the rows.

Edit: Looking back I see that this solution is quite similar to Luisi's. Should have looked before I leaped.