I have a dataset that I have aggregated at monthly level. The next part needs me to take, for every block of 3 months, the sum of the data at monthly level.
So essentially my input data (after aggregated to monthly level) looks like:
month
year
status
count_id
08
2021
stat_1
1
09
2021
stat_1
3
10
2021
stat_1
5
11
2021
stat_1
10
12
2021
stat_1
10
01
2022
stat_1
5
02
2022
stat_1
20
and then my output data to look like:
month
year
status
count_id
3m_sum
08
2021
stat_1
1
1
09
2021
stat_1
3
4
10
2021
stat_1
5
8
11
2021
stat_1
10
18
12
2021
stat_1
10
25
01
2022
stat_1
5
25
02
2022
stat_1
20
35
i.e 3m_sum for Feb = Feb + Jan + Dec. I tried to do this using a self join and wrote a query along the lines of
WITH CTE AS(
SELECT date_part('month',date_col) as month
,date_part('year',date_col) as year
,status
,count(distinct id) as count_id
FROM (date_col, status, transaction_id) as a
)
SELECT a.month, a.year, a.status, sum(b.count_id) as 3m_sum
from cte as a
left join cte as b on a.status = b.status
and b.month >= a.month - 2 and b.month <= a.month
group by 1,2,3
This query NEARLY works. Where it falls apart is in Jan and Feb. My data is from August 2021 to Apr 2022. The means, the value for Jan should be Nov + Dec + Jan. Similarly for Feb it should be Dec + Jan + Feb.
As I am doing a join on the MONTH, all the months of Aug - Nov are treated as being values > month of jan/feb and so the query isn't doing the correct sum.
How can I adjust this bit to give the correct sum?
I did think of using a LAG function, but (even though I'm 99% sure a month won't ever be missed), I can't guarantee we will never have a month with 0 values, and therefore my LAG function will be summing the wrong rows.
I also tried doing the same join, but at individual date level (and not aggregating in my nested query) but this gave vastly different numbers, as I want the sum of the aggregation and I think the sum from the individual row was duplicated a lot of stuff I do a COUNT DISTINCT on to remove.
You can use a SUM with a window frame of 2 PRECEDING. To ensure you don't miss rows, use a calendar table and left-join all the results to it.
SELECT *,
SUM(a.count_id) OVER (ORDER BY c.year, c.month ROWS BETWEEN 2 PRECEDING AND CURRENT ROW)
FROM Calendar c
LEFT JOIN a ON a.year = c.year AND a.month = c.month
WHERE c.year >= 2021 AND c.year <= 2022;
db<>fiddle
You could also use LAG but you would need it twice.
It should be #Charlieface's answer - only that I get one different result than you put in your expected result table:
WITH
-- your input - and I avoid keywords like "MONTH" or "YEAR"
-- and also identifiers starting with digits are forbidden -
indata(mm,yy,status,count_id,sum_3m) AS (
SELECT 08,2021,'stat_1',1,1
UNION ALL SELECT 09,2021,'stat_1',3,4
UNION ALL SELECT 10,2021,'stat_1',5,8
UNION ALL SELECT 11,2021,'stat_1',10,18
UNION ALL SELECT 12,2021,'stat_1',10,25
UNION ALL SELECT 01,2022,'stat_1',5,25
UNION ALL SELECT 02,2022,'stat_1',20,35
)
SELECT
*
, SUM(count_id) OVER(
ORDER BY yy,mm
ROWS BETWEEN 2 PRECEDING AND CURRENT ROW
) AS sum_3m_calc
FROM indata;
-- out mm | yy | status | count_id | sum_3m | sum_3m_calc
-- out ----+------+--------+----------+--------+-------------
-- out 8 | 2021 | stat_1 | 1 | 1 | 1
-- out 9 | 2021 | stat_1 | 3 | 4 | 4
-- out 10 | 2021 | stat_1 | 5 | 8 | 9
-- out 11 | 2021 | stat_1 | 10 | 18 | 18
-- out 12 | 2021 | stat_1 | 10 | 25 | 25
-- out 1 | 2022 | stat_1 | 5 | 25 | 25
-- out 2 | 2022 | stat_1 | 20 | 35 | 35
Have a table with the following schema design and the data residing inside it is like:
ID HITS MISS DDATE
1 10 3 20180101
1 33 21 20180122
1 84 11 20180901
1 11 2 20180405
1 54 23 20190203
1 33 43 20190102
4 54 22 20170305
4 56 88 20180115
5 87 22 20180809
5 66 48 20180617
5 91 53 20170606
DataTypes:
ID INT
HITS INT
MISS INT
DDATE STRING
The requirement is to calculate the total of the given (HITS and MISS) on yearly basis i.e 2017,2018,2019...
Written the following query:
SELECT ID,
SUM(HITS) AS HITS,SUM(MISS) AS MISS,
CASE
WHEN DDATE BETWEEN '201701' AND '201712' THEN '2017' ELSE
'NOTHING' END AS TTL_YR17_DATA
CASE
WHEN DDATE BETWEEN '201801' AND '201812' THEN '2018' ELSE
'NOTHING' END AS TTL_YR18_DATA
CASE
WHEN DDATE BETWEEN '201901' AND '201912' THEN '2019' ELSE
'NOTHING' END AS TTL_YR19_DATA
FROM
HST_TABLE
WHERE
DDATE BETWEEN '201801' AND '201812'
GROUP BY
ID,DDATE;
But, the query is not fetching the expected result.
Actual O/P:
1 10 3 2018
1 33 21 2018
1 84 11 2018
1 11 2 2018
1 54 23 2019
1 33 43 2019
4 54 22 2017
4 56 88 2018
5 87 22 2018
5 66 48 2018
5 91 53 2017
Expected O/P:
1 138 37 2018
4 56 88 2018
5 153 70 2018
1 87 66 2019
5 91 53 2017
Another related question:
Is there a way that I can avoid passing the DDATE range in the query? As this should be given by the user and shouldn't be hardcoded.
Any help/advice to achieve the above two requirements will be really helpful.
OK,it's easy to implement this with the substring function in HIVE, as below:
select
substring(dddate,0,4) as the_year,
id,
sum(hits) as hits_num,
sum(miss) as miss_num
from
hst_table
group by
substring(dddate,0,4),
id
order by
the_year,
id
The answer above by #Shawn.X is correct but has a logical flaw. Below is the corrected one:
select
substring(ddate,0,4) as the_year,
id,
sum(hits) as hits_num,
sum(miss) as miss_num
from
hst_table
group by
substring(ddate,0,4),
id
order by
the_year,
id;
I have a table where for some dates a certain number of entries are placed. Here is the table structure :
ID EntryName Entries DateOfEntry
1 A 20 2016-01-17
2 B 22 2016-01-29
3 C 23 2016-02-17
4 D 19 2016-02-17
5 E 29 2016-03-17
6 F 30 2016-03-17
7 G 43 2016-04-17
8 H 10 2016-04-17
9 I 5 2016-05-17
10 J 120 2016-05-17
11 K 220 2016-06-17
12 L 210 2016-06-17
13 M 10 2016-07-17
14 N 20 2016-07-17
15 O 15 2016-08-17
16 P 17 2016-08-17
17 Q 19 2016-09-17
18 R 23 2016-09-17
19 S 43 2016-10-17
20 T 56 2016-10-17
21 U 65 2016-11-17
22 V 78 2016-11-17
23 W 12 2016-12-17
24 X 23 2016-12-17
25 Y 43 2016-02-17
26 Z 67 2016-03-17
27 AA 35 2015-01-17
28 AB 23 2015-01-29
29 AC 43 2015-02-17
30 AD 35 2015-02-17
31 AE 45 2015-03-17
32 AF 23 2015-03-17
33 AG 43 2015-04-17
34 AH 19 2015-04-17
35 AI 21 2015-05-17
36 AJ 13 2015-05-17
37 AK 22 2015-06-17
38 AL 45 2015-06-17
39 AM 66 2015-07-17
40 AN 77 2015-07-17
41 AO 89 2015-08-17
42 AP 127 2015-08-17
43 AQ 19 2015-09-17
44 AR 223 2015-09-17
45 AS 143 2015-10-17
46 AT 36 2015-10-17
47 AU 45 2015-11-17
48 AV 28 2015-11-17
49 AW 72 2015-12-17
50 AX 24 2015-12-17
51 AY 46 2015-02-17
52 AZ 62 2015-03-17
The column EntryName is the entry identifier, the column Entries has the total number of entries for the date specified in the column DateOfEntry.
I am trying to formulate a query where the total number of entries are displayed on a month-wise basis. I currently have this query :
SELECT DateName(MONTH, e.DateOfEntry) AS MonthOfEntry,
MONTH(e.DateOfEntry) AS MonthNumber,
SUM(e.Entries) AS TotalEntries
FROM #entry e
GROUP BY MONTH(e.DateOfEntry), DateName(MONTH,e.DateOfEntry)
ORDER BY MONTH(e.DateOfEntry) ASC
which works fine as far as displaying the results are concerned. However, my issue here is that I need to sort the results on a month-wise basis where the starting month would be dynamic i.e. arising from a parameter (supplied by the user).
This means that if the user selects May of 2015 the results should be sorted from May 2015 to April 2016. Similarly, if the user selects October 2015, the results would be displayed from October 2015 to September 2016.
How would I go about getting this condition within the ORDER BY clause ?
You can put an offset into the ORDER BY using modulo arithmetic. For April:
ORDER BY (MONTH(e.DateOfEntry) + 12 - 4) % 12
--------------------------------------^ month number to start with
(The + 12 is simply so I don't have to remember if % returns negative numbers with negative operands.)
If you want the results chronologically, you can instead do:
ORDER BY MIN(e.DateOfEntry)
You could use the belosw in order by
ORDER BY YEAR(e.DATEOFENTRY),
DATEPART(MM,e.DAREOFENTRY)
This will sort the result first for Year and next for month.
Here you need to specify these same columns in Select.
If I understood you correctly
"This means that if the user selects May of 2015 the results should be sorted from May 2015 to April 2016. Similarly, if the user selects October 2015, the results would be displayed from October 2015 to September 2016."
this should work:
SAMPLE DATA:
IF OBJECT_ID('tempdb..#entry') IS NOT NULL
DROP TABLE #entry;
CREATE TABLE #entry(ID INT ,EntryName VARCHAR(10) , Entries INT , DateOfEntry DATE);
INSERT INTO #entry (ID ,EntryName ,Entries ,DateOfEntry)
VALUES
(1 ,'A', 20 ,'2016-01-17'),
(2 ,'B', 22 ,'2016-01-29'),
(3 ,'C', 23 ,'2016-02-17'),
(4 ,'D', 19 ,'2016-02-17'),
(5 ,'E', 29 ,'2016-03-17'),
(6 ,'F', 30 ,'2016-03-17'),
(7 ,'G', 43 ,'2016-04-17'),
(8 ,'H', 10 ,'2016-04-17'),
(9 ,'I', 5 ,'2016-05-17'),
(10,'J', 120 ,'2016-05-17'),
(11,'K', 220 ,'2016-06-17'),
(12,'L', 210 ,'2016-06-17'),
(13,'M', 10 ,'2016-07-17'),
(14,'N', 20 ,'2016-07-17'),
(15,'O', 15 ,'2016-08-17'),
(16,'P', 17 ,'2016-08-17'),
(17,'Q', 19 ,'2016-09-17'),
(18,'R', 23 ,'2016-09-17'),
(19,'S', 43 ,'2016-10-17'),
(20,'T', 56 ,'2016-10-17'),
(21,'U', 65 ,'2016-11-17'),
(22,'V', 78 ,'2016-11-17'),
(23,'W', 12 ,'2016-12-17'),
(24,'X', 23 ,'2016-12-17'),
(25,'Y', 43 ,'2016-02-17'),
(26,'Z', 67 ,'2016-03-17'),
(27,'AA',35 ,'2015-01-17'),
(28,'AB',23 ,'2015-01-29'),
(29,'AC',43 ,'2015-02-17'),
(30,'AD',35 ,'2015-02-17'),
(31,'AE',45 ,'2015-03-17'),
(32,'AF',23 ,'2015-03-17'),
(33,'AG',43 ,'2015-04-17'),
(34,'AH',19 ,'2015-04-17'),
(35,'AI',21 ,'2015-05-17'),
(36,'AJ',13 ,'2015-05-17'),
(37,'AK',22 ,'2015-06-17'),
(38,'AL',45 ,'2015-06-17'),
(39,'AM',66 ,'2015-07-17'),
(40,'AN',77 ,'2015-07-17'),
(41,'AO',89 ,'2015-08-17'),
(42,'AP',127 ,'2015-08-17'),
(43,'AQ',19 ,'2015-09-17'),
(44,'AR',223 ,'2015-09-17'),
(45,'AS',143 ,'2015-10-17'),
(46,'AT',36 ,'2015-10-17'),
(47,'AU',45 ,'2015-11-17'),
(48,'AV',28 ,'2015-11-17'),
(49,'AW',72 ,'2015-12-17'),
(50,'AX',24 ,'2015-12-17'),
(51,'AY',46 ,'2015-02-17'),
(52,'AZ',62 ,'2015-03-17')
QUERY WITH PARAMS:
DECLARE #Month VARCHAR(2) = '05', #Year VARCHAR(4) = '2015'
SELECT DateName(MONTH, e.DateOfEntry) AS MonthOfEntry,
MONTH(e.DateOfEntry) AS MonthNumber,
SUM(e.Entries) AS TotalEntries
FROM #entry e
WHERE CAST(e.DateOfEntry AS DATE) >= CAST( #Year+#Month+'01' AS DATE)
GROUP BY MONTH(e.DateOfEntry), DateName(MONTH,e.DateOfEntry)
ORDER BY MONTH(e.DateOfEntry) ASC
RESULTS:
add a where clause for the query
WHERE MONTH(e.DateOfEntry) < User.Month AND YEAR(e.DateOfEntry) < User.Year AND MONTH(e.DateOfEntry) > (User.Month-1) AND YEAR(e.DateOfEntry) > (User.Year+1)
Assuming your parameter is an Integer called #FirstMonth, you could get the proper month order using:
Case
WHEN MONTH(e.DateOfEntry) < #FirstMonth then MONTH(e.DateOfEntry) + 12
ELSE MONTH(e.DateOfEntry)
END AS MonthNumber
Of all the answers and suggestions I have come across here, I find this way (suggested by xQbert in the question's comments) to be the simplest one :
SELECT DateName(MONTH, e.DateOfEntry) + ' ' + CONVERT(NVARCHAR(100), YEAR(e.DateOfEntry)) AS MonthOfEntry,
MONTH(e.DateOfEntry) AS MonthNumber,
SUM(e.Entries) AS TotalEntries
FROM Entry e
WHERE e.DateOfEntry BETWEEN #StartDate AND (DATEADD(YEAR, 1, #StartDate))
GROUP BY MONTH(e.DateOfEntry), DateName(MONTH,e.DateOfEntry), YEAR(e.DateOfEntry)
ORDER BY YEAR(e.DateOfEntry) ASC, MONTH(e.DateOfEntry) ASC
A fiddle to demonstrate this : http://rextester.com/CJFFP5640
Initially, I was using the following query :
SELECT sortingList.MonthOfEntry,
sortingList.TotalEntries,
sortingList.MonthNumber
FROM (
SELECT DateName(MONTH, DATEADD(MONTH, MONTH(e.DateOfEntry), 0) - 1) + ' ' + CONVERT(nvarchar(20),YEAR(e.DateOfEntry)) AS MonthOfEntry,
SUM(e.Entries) as TotalEntries,
CASE
WHEN ((MONTH(e.DateOfEntry) - MONTH(#StartDate)) > 0)
THEN (MONTH(e.DateOfEntry) - MONTH(#StartDate)) + 1
WHEN ((MONTH(e.DateOfEntry) - MONTH(#StartDate)) = 0)
THEN 1
ELSE
((12 - MONTH(#StartDate)) + (MONTH(e.DateOfEntry))) + 1
END
AS MonthNumber
FROM Entry e
WHERE e.DateOfEntry >= #StartDate AND e.DateOfEntry < DATEADD(YEAR, 1, #StartDate)
GROUP BY DateName(MONTH, DATEADD(MONTH, MONTH(e.DateOfEntry), 0) - 1), YEAR(e.DateOfEntry), MONTH(e.DateOfEntry) - MONTH(#StartDate), MONTH(e.DateOfEntry)
) sortingList
ORDER BY sortingList.MonthNumber ASC
Here's an Fiddle to demonstrate this : http://rextester.com/LEVD30653
Explanation (non TL;DR)
You can see that it's essentially the same WHERE clause. However, the query at the top uses much simpler logic for sorting and is more fluent and readable.
Do note that the second solution (using the CASE statement) sorts the month numbers as per the user-provided month number i.e. if the user provides December 2015, then the second solution will number Dec 2015 as 1, January 2016 as 2, February 2016 as 3 and so on and so forth. This might be more beneficial in cases where you want to work on top of this data.
As far as my use-case is concerned, this makes more sense. However, as far as the scope of the question is concerned, the query at the top is the best one.