I have a table with emplid and end_date columns. I want from all emplids the max end_dates. If at least one end_date is null, I want to have the null value as max. So in this example:
emplid end_date
1 05/04/2019
1 05/10/2019
1 null
2 05/04/2019
2 05/10/2019
I want as result:
emplid end_date
1 null
2 05/10/2019
I tried something like
select emplid,
CASE
WHEN MAX(NVL(end_Date,'01/01/3000'))='01/01/3000' THEN null
ELSE end_date
END as end_dt
from people
group by emplid
then I get a group-by error.
Maybe it is very easy, but I don't figure out how to get properly what I want.
with s(id, dt) as (
select 1, to_date('05/04/2019', 'dd/mm/yyyy') from dual union all
select 1, to_date('05/10/2019', 'dd/mm/yyyy') from dual union all
select 1, null from dual union all
select 2, to_date('05/04/2019', 'dd/mm/yyyy') from dual union all
select 2, to_date('05/10/2019', 'dd/mm/yyyy') from dual)
select id, decode(count(dt), count(*), max(dt)) max_dt
from s
group by id;
ID MAX_DT
---------- -----------------------------
1
2 2019-10-05 00:00:00
I would simply do:
select emplid,
(case when count(*) = count(end_date)
then max(end_date)
end) as max_end_date
from t
group by emplid;
There is no reason to introduce a "magic" maximum value (even if it is correct).
The first expression in the case is simply asking "do the number of non-NULL end-date values match the number of rows".
Try this
SELECT
EMPLID,
CASE WHEN END_DATE='01/01/3000' THEN NULL ELSE END_DATE END AS END_DT
FROM
(
SELECT EMPLID, MAX(END_DATE) AS END_DATE FROM
(
SELECT EMPLID, NVL(END_DATE,'01/01/3000') AS END_DATE FROM PEOPLE
)
GROUP BY EMPLID
);
Case does not go with group by , you have to get the max value using group by first then evaluate the null values. Try below.
select empid, CASE WHEN NVL(eDate,'01-DEC-3000')='01-DEC-3000' THEN null ELSE edate end end_dt from (
select empid, MAX(NVL(eDate,'01-DEC-3000')) eDate
from
(select 1 empid, sysdate-100 edate from dual union all
select 1 empid, sysdate-10 edate from dual union all
select 1 empid, null edate from dual union all
select 2 empid, sysdate-105 edate from dual union all
select 2 empid, sysdate-1 edate from dual ) datad
group by empid);
Related
I'm trying to build a query for the following scenario,
Group records by license ID and get min and max dates
For a given license ID, if there are two earliest start dates, then start date of the particular ID has to be updated as latest start date in that grouping.
Since I'm new to sql, I need help to satisfy condition 2. Any help is greatly appreciated. Thanks
Actual data
LicenseID
StartDate
EndDate
100
4/3/2000
3/1/2013
100
4/3/2000
2/2/2017
100
3/1/2013
1/23/2015
100
1/23/2015
2/2/2017
100
2/2/2017
2/9/2018
100
2/2/2017
12/18/2018
100
12/18/2018
2/16/2021
Expected output
LicenseID
StartDate
EndDate
100
12/18/2018
2/16/2021
Here's one option; read comments within code.
Sample data:
SQL> with test (id, start_date, end_date) as
2 (select 100, date '2000-04-03', date '2013-03-01' from dual union all
3 select 100, date '2000-04-03', date '2017-02-02' from dual union all
4 select 100, date '2018-12-18', date '2021-02-16' from dual
5 ),
Query begins here:
6 -- rank start dates per each ID
7 temp as
8 (select id,
9 min(start_date) over (partition by id) min_sd,
10 max(start_date) over (partition by id) max_sd,
11 rank() over (partition by id order by start_date) rnk_sd,
12 --
13 max(end_date) over (partition by id) max_ed
14 from test
15 ),
16 -- count number of the 1st start dates
17 temp2 as
18 (select id,
19 sum(case when rnk_sd = 1 then 1 else 0 end) cnt_sd
20 from temp
21 group by id
22 )
23 -- if number of the 1st start dates is 1, take MIN_SD. Otherwise, take MAX_SD
24 select distinct
25 b.id,
26 case when b.cnt_sd = 1 then a.min_sd else a.max_sd end start_date,
27 a.max_ed end_date
28 from temp2 b join temp a on a.id = b.id;
Result:
ID START_DATE END_DATE
---------- ---------- ----------
100 12/18/2018 02/16/2021
SQL>
This can filter them:
WITH sample_data AS
(
SELECT 100 AS LicenseID, TO_DATE('04/03/2000','MM/DD/YYYY') AS StartDate, TO_DATE('03/01/2013','MM/DD/YYYY') AS EndDate FROM DUAL UNION ALL
SELECT 100, TO_DATE('04/03/2000','MM/DD/YYYY'), TO_DATE('02/02/2017','MM/DD/YYYY') FROM DUAL UNION ALL
SELECT 100, TO_DATE('03/01/2013','MM/DD/YYYY'), TO_DATE('01/23/2015','MM/DD/YYYY') FROM DUAL UNION ALL
SELECT 100, TO_DATE('01/23/2015','MM/DD/YYYY'), TO_DATE('02/02/2017','MM/DD/YYYY') FROM DUAL UNION ALL
SELECT 100, TO_DATE('02/02/2017','MM/DD/YYYY'), TO_DATE('02/09/2018','MM/DD/YYYY') FROM DUAL UNION ALL
SELECT 100, TO_DATE('02/02/2017','MM/DD/YYYY'), TO_DATE('12/18/2018','MM/DD/YYYY') FROM DUAL UNION ALL
SELECT 100, TO_DATE('12/18/2018','MM/DD/YYYY'), TO_DATE('02/16/2021','MM/DD/YYYY') FROM DUAL
)
SELECT dat.licenseID, CASE WHEN dups.licenseID IS NOT NULL THEN MAX(StartDate)
ELSE MIN(StartDate)
END,
CASE WHEN dups.licenseID IS NOT NULL THEN MAX(EndDate)
ELSE MIN(EndDate)
END
FROM sample_data dat
LEFT OUTER JOIN (SELECT COUNT(1), sd.LicenseID
FROM sample_data sd
INNER JOIN (SELECT MIN(StartDate) AS StartDate, LicenseID
FROM sample_data
GROUP BY LicenseID) mins
ON sd.LicenseID = mins.LicenseID AND sd.startDate = mins.StartDate
GROUP BY sd.LicenseID
HAVING COUNT(1) > 1) dups
ON dups.LicenseID = dat.licenseID
GROUP BY dat.licenseID, dups.licenseID;
You can use:
SELECT licenseid,
MAX(startdate) AS startdate,
MAX(enddate) KEEP (DENSE_RANK LAST ORDER BY startdate) AS enddate
FROM table_name
GROUP BY licenseid
HAVING COUNT(*) KEEP (DENSE_RANK FIRST ORDER BY startdate) > 1;
or:
SELECT licenseid,
max_startdate AS startdate,
max_enddate As enddate
FROM (
SELECT licenseid,
RANK()
OVER (PARTITION BY licenseid ORDER BY startdate) AS rnk,
ROW_NUMBER()
OVER (PARTITION BY licenseid, startdate ORDER BY enddate) AS rn,
MAX(startdate)
OVER (PARTITION BY licenseid) AS max_startdate,
MAX(enddate)
KEEP (DENSE_RANK LAST ORDER BY startdate)
OVER (PARTITION BY licenseid) AS max_enddate
FROM table_name t
)
WHERE rnk = 1
AND rn = 2;
Which, for the sample data:
CREATE TABLE table_name (licenseid, startdate, enddate) AS
SELECT 100, DATE'2000-04-03', DATE'2013-03-01' FROM DUAL UNION ALL
SELECT 100, DATE'2000-04-03', DATE'2017-02-02' FROM DUAL UNION ALL
SELECT 100, DATE'2013-03-01', DATE'2015-01-23' FROM DUAL UNION ALL
SELECT 100, DATE'2015-01-23', DATE'2017-02-02' FROM DUAL UNION ALL
SELECT 100, DATE'2017-02-02', DATE'2018-02-09' FROM DUAL UNION ALL
SELECT 100, DATE'2018-02-02', DATE'2018-12-18' FROM DUAL UNION ALL
SELECT 100, DATE'2018-12-18', DATE'2021-02-16' FROM DUAL;
Both output:
LICENSEID
STARTDATE
ENDDATE
100
2018-12-18 00:00:00
2021-02-16 00:00:00
If you do want to perform an UPDATE of that second row then:
MERGE INTO table_name dst
USING (
SELECT ROWID AS rid,
max_startdate,
max_enddate
FROM (
SELECT RANK()
OVER (PARTITION BY licenseid ORDER BY startdate) AS rnk,
ROW_NUMBER()
OVER (PARTITION BY licenseid, startdate ORDER BY enddate) AS rn,
MAX(startdate)
OVER (PARTITION BY licenseid) AS max_startdate,
MAX(enddate)
KEEP (DENSE_RANK LAST ORDER BY startdate)
OVER (PARTITION BY licenseid) AS max_enddate
FROM table_name t
)
WHERE rnk = 1
AND rn = 2
)src
ON (src.rid = dst.ROWID)
WHEN MATCHED THEN
UPDATE
SET startdate = src.max_startdate,
enddate = src.max_enddate;
db<>fiddle here
The problem I am facing is how to find distinct time periods from multiple time periods with overlap in Teradata ANSI SQL.
For example, the attached tables contain multiple overlapping time periods, how can I combine those time periods into 3 unique time periods in Teradata SQL???
I think I can do it in python with the loop function, but not sure how to do it in SQL
ID
Start Date
End Date
001
2005-01-01
2006-01-01
001
2005-01-01
2007-01-01
001
2008-01-01
2008-06-01
001
2008-04-01
2008-12-01
001
2010-01-01
2010-05-01
001
2010-04-01
2010-12-01
001
2010-11-01
2012-01-01
My expected result is:
ID
start_Date
end_date
001
2005-01-01
2007-01-01
001
2008-01-01
2008-12-01
001
2010-01-01
2012-01-01
From Oracle 12, you can use MATCH_RECOGNIZE to perform a row-by-row comparison:
SELECT *
FROM table_name
MATCH_RECOGNIZE(
PARTITION BY id
ORDER BY start_date
MEASURES
FIRST(start_date) AS start_date,
MAX(end_date) AS end_date
ONE ROW PER MATCH
PATTERN (overlapping_ranges* last_range)
DEFINE overlapping_ranges AS NEXT(start_date) <= MAX(end_date)
)
Which, for the sample data:
CREATE TABLE table_name (ID, Start_Date, End_Date) AS
SELECT '001', DATE '2005-01-01', DATE '2006-01-01' FROM DUAL UNION ALL
SELECT '001', DATE '2005-01-01', DATE '2007-01-01' FROM DUAL UNION ALL
SELECT '001', DATE '2008-01-01', DATE '2008-06-01' FROM DUAL UNION ALL
SELECT '001', DATE '2008-04-01', DATE '2008-12-01' FROM DUAL UNION ALL
SELECT '001', DATE '2010-01-01', DATE '2010-05-01' FROM DUAL UNION ALL
SELECT '001', DATE '2010-04-01', DATE '2010-12-01' FROM DUAL UNION ALL
SELECT '001', DATE '2010-11-01', DATE '2012-01-01' FROM DUAL;
Outputs:
ID
START_DATE
END_DATE
001
2005-01-01 00:00:00
2007-01-01 00:00:00
001
2008-01-01 00:00:00
2008-12-01 00:00:00
001
2010-01-01 00:00:00
2012-01-01 00:00:00
db<>fiddle here
Update: Alternative query
SELECT id,
start_date,
end_date
FROM (
SELECT id,
dt,
SUM(cnt) OVER (PARTITION BY id ORDER BY dt) AS grp,
cnt
FROM (
SELECT ID,
dt,
SUM(type) OVER (PARTITION BY id ORDER BY dt, ROWNUM) * type AS cnt
FROM table_name
UNPIVOT (dt FOR type IN (start_date AS 1, end_date AS -1))
)
WHERE cnt IN (1,0)
)
PIVOT (MAX(dt) FOR cnt IN (1 AS start_date, 0 AS end_date))
Or, an equivalent that does not use UNPIVOT, PIVOT or ROWNUM and works in both Oracle and PostgreSQL:
SELECT id,
MAX(CASE cnt WHEN 1 THEN dt END) AS start_date,
MAX(CASE cnt WHEN 0 THEN dt END) AS end_date
FROM (
SELECT id,
dt,
SUM(cnt) OVER (PARTITION BY id ORDER BY dt) AS grp,
cnt
FROM (
SELECT ID,
dt,
SUM(type) OVER (PARTITION BY id ORDER BY dt, rn) * type AS cnt
FROM (
SELECT r.*,
ROW_NUMBER() OVER (PARTITION BY id ORDER BY dt ASC, type DESC) AS rn
FROM (
SELECT id, 1 AS type, start_date AS dt FROM table_name
UNION ALL
SELECT id, -1 AS type, end_date AS dt FROM table_name
) r
) p
) s
WHERE cnt IN (1,0)
) t
GROUP BY id, grp
Update 2: Another Alternative
SELECT id,
MIN(start_date) AS start_date,
MAX(end_Date) AS end_date
FROM (
SELECT t.*,
SUM(CASE WHEN start_date <= prev_max THEN 0 ELSE 1 END)
OVER (PARTITION BY id ORDER BY start_date) AS grp
FROM (
SELECT t.*,
MAX(end_date) OVER (
PARTITION BY id ORDER BY start_date
ROWS BETWEEN UNBOUNDED PRECEDING AND 1 PRECEDING
) AS prev_max
FROM table_name t
) t
) t
GROUP BY id, grp
db<>fiddle Oracle PostgreSQL
This is a gaps and islands problem. Try this:
with u as
(select ID, start_date, end_date,
case
when start_date <= lag(end_date) over(partition by ID order by start_date, end_date) then 0
else 1 end as grp
from table_name),
v as
(select ID, start_date, end_date,
sum(grp) over(partition by ID order by start_date, end_date) as island
from u)
select ID, min(start_date) as start_Date, max(end_date) as end_date
from v
group by ID, island;
Fiddle
Basically you can identify "islands" by comparing start_date of current row to end_date of previous row (ordered by start_date, end_date), if it precedes it then it's the same island. Then you can do a rolling sum() to get the island numbers. Finally select min(start_date) and max(end_date) from each island to get the desired output.
This may work ,with little bit of change in function , I tried it in Dbeaver :
select ID,Start_Date,End_Date
from
(
select t.*,
dense_rank () over(partition by extract (year from Start_Date) order BY End_Date desc) drnk
from testing_123 t
) temp
where temp.drnk = 1
ORDER BY Start_Date;
Try this
WITH a as (
SELECT
ID,
LEFT(Start_Date, 4) as Year,
MIN(Start_Date) as New_Start_Date
FROM
TAB1
GROUP BY
ID,
LEFT(Start_Date, 4)
), b as (
SELECT
a.ID,
Year,
New_Start_Date,
End_Date
FROM
a
LEFT JOIN
TAB1
ON LEFT(a.New_Start_Date, 4) = LEFT(TAB1.Start_Date, 4)
)
select
ID,
New_Start_Date as Start_Date,
MAX(End_Date)
from
b
GROUP BY
ID,
New_Start_Date;
Example: https://dbfiddle.uk/?rdbms=mysql_8.0&fiddle=97f91b68c635aebfb752538cdd752ace
I want to look at the lead type and if that type is the same for that row then merge in those dates to fit within one row.
I have the below table:
id start_dt end_dt type
1 1/1/19 2/21/19 cross
1 2/22/19 6/5/19 cross
1 6/6/19 8/31/19 cross
1 9/1/19 10/3/19 AAAA
1 10/4/19 10/4/19 cross
1 10/5/19 10/6/19 AAAA
1 10/7/19 10/10/19 AAAA
1 10/11/19 12/31/99 cross
Expected Results:
id start_dt end_dt type
1 1/1/19 8/31/19 cross
1 9/1/19 10/3/19 AAAA
1 10/4/19 10/4/19 cross
1 10/5/19 10/10/19 AAAA
1 10/11/19 12/31/99 cross
How can I get my output to look like the expected results?
I have tested withlead lag rank and case expression but nothing worthy of adding here. Am I on the right path?
This is a gaps-and-islands problem. One option for solving it through contribution of row_number() analytical function :
select min(start_dt) as startdate, max(end_dt) as enddate, type
from
(
with t(id, start_dt, end_dt,type) as
(
select 1, date'2019-01-01', date'2019-02-21', 'cross' from dual union all
select 1, date'2019-02-22', date'2019-06-05', 'cross' from dual union all
select 1, date'2019-06-06', date'2019-08-31', 'cross' from dual union all
select 1, date'2019-09-01', date'2019-10-03', 'AAAA' from dual union all
select 1, date'2019-09-04', date'2019-10-04', 'cross' from dual union all
select 1, date'2019-10-05', date'2019-10-06', 'AAAA' from dual union all
select 1, date'2019-10-07', date'2019-10-10', 'AAAA' from dual union all
select 1, date'2019-10-11', date'2019-12-31', 'cross' from dual
)
select type,
row_number() over (partition by id, type order by end_dt) as rn1,
row_number() over (partition by id order by end_dt) as rn2,
start_dt, end_dt
from t
) tt
group by type, rn1 - rn2
order by enddate;
STARTDATE ENDDATE TYPE
--------- --------- -----
01-JAN-19 31-AUG-19 cross
01-SEP-19 03-OCT-19 AAAA
04-SEP-19 04-OCT-19 cross
05-OCT-19 10-OCT-19 AAAA
11-OCT-19 31-DEC-19 cross
Demo
I actually think this is a pretty good case for Oracle's Pattern Matching Functionality.
with t(id, start_dt, end_dt,type) as
(
select 1, date'2019-01-01', date'2019-02-21', 'cross' from dual union all
select 1, date'2019-02-22', date'2019-06-05', 'cross' from dual union all
select 1, date'2019-06-06', date'2019-08-31', 'cross' from dual union all
select 1, date'2019-09-01', date'2019-10-03', 'AAAA' from dual union all
select 1, date'2019-09-04', date'2019-10-04', 'cross' from dual union all
select 1, date'2019-10-05', date'2019-10-06', 'AAAA' from dual union all
select 1, date'2019-10-07', date'2019-10-10', 'AAAA' from dual union all
select 1, date'2019-10-11', date'2019-12-31', 'cross' from dual
)
SELECT *
FROM t
MATCH_RECOGNIZE(ORDER BY start_dt
MEASURES a.id AS ID,
A.start_dt AS START_DT,
NVL(LAST(B.end_dt), A.end_dt) AS END_DT,
a.type AS TYPE
PATTERN (A B*)
DEFINE B AS start_dt > PREV(start_dt) AND type = PREV(type));
A detailed primer on the topic can be found here
If you want to look at adjacent rows to find groups that can combine, then I recommend lag() to find where groups start and a cumulative sum on that:
select id, type, min(start_dt), max(end_dt)
from (select t.*,
sum(case when prev_end_dt >= start_dt - 1 then 0 else 1 end) over (partition by id, type order by start_dt) as grp
from (select t.*,
lag(end_dt) over (partition by id, type order by start_dt) as prev_end_dt
from t
) t
) t
group by id, type, grp
order by id, min(start_dt);
In particular, this will find cases where the type does not change but there is a gap in the time frames, as shown by this db<>fiddle for id = 2.
I have a table named x . The data is as follows.
Acccount_num start_dt end_dt
A111326 02/01/2016 02/11/2016
A111326 02/12/2016 03/05/2016
A111326 03/02/2016 03/16/2016
A111331 02/28/2016 02/29/2016
A111331 02/29/2016 03/29/2016
A999999 08/25/2015 08/25/2015
A999999 12/19/2015 12/22/2015
A222222 11/06/2015 11/10/2015
A222222 05/16/2016 05/17/2016
Both A111326 and A111331 should be identified as contiguous data and A999999 and
A222222 should be identified as discontinuous data.In my code I currently use the following query to identify discontinuous data. The A111326 is also erroneously identified as discontinuous data. Please help to modify the below code so that A111326 is not identified as discontinuous data.Thanks in advance for your help.
(SELECT account_num
FROM (SELECT account_num,
(MAX (
END_DT)
OVER (PARTITION BY account_num
ORDER BY START_DT))
START_DT,
(LEAD (
START_DT)
OVER (PARTITION BY account_num
ORDER BY START_DT))
END_DT
FROM x
WHERE (START_DT + 1) <=
(END_DT - 1))
WHERE START_DT < END_DT);
Oracle Setup:
CREATE TABLE accounts ( Account_num, start_dt, end_dt ) AS
SELECT 'A', DATE '2016-02-01', DATE '2016-02-11' FROM DUAL UNION ALL
SELECT 'A', DATE '2016-02-12', DATE '2016-03-05' FROM DUAL UNION ALL
SELECT 'A', DATE '2016-03-02', DATE '2016-03-16' FROM DUAL UNION ALL
SELECT 'B', DATE '2016-02-28', DATE '2016-02-29' FROM DUAL UNION ALL
SELECT 'B', DATE '2016-02-29', DATE '2016-03-29' FROM DUAL UNION ALL
SELECT 'C', DATE '2015-08-25', DATE '2015-08-25' FROM DUAL UNION ALL
SELECT 'C', DATE '2015-12-19', DATE '2015-12-22' FROM DUAL UNION ALL
SELECT 'D', DATE '2015-11-06', DATE '2015-11-10' FROM DUAL UNION ALL
SELECT 'D', DATE '2016-05-16', DATE '2016-05-17' FROM DUAL UNION ALL
SELECT 'E', DATE '2016-01-01', DATE '2016-01-02' FROM DUAL UNION ALL
SELECT 'E', DATE '2016-01-05', DATE '2016-01-06' FROM DUAL UNION ALL
SELECT 'E', DATE '2016-01-03', DATE '2016-01-07' FROM DUAL;
Query:
WITH times ( account_num, dt, lvl ) AS (
SELECT Account_num, start_dt - 1, 1 FROM accounts
UNION ALL
SELECT Account_num, end_dt, -1 FROM accounts
)
, totals ( account_num, dt, total ) AS (
SELECT account_num,
dt,
SUM( lvl ) OVER ( PARTITION BY Account_num ORDER BY dt, lvl DESC )
FROM times
)
SELECT Account_num,
CASE WHEN COUNT( CASE total WHEN 0 THEN 1 END ) > 1
THEN 'N'
ELSE 'Y'
END AS is_contiguous
FROM totals
GROUP BY Account_Num
ORDER BY Account_Num;
Output:
ACCOUNT_NUM IS_CONTIGUOUS
----------- -------------
A Y
B Y
C N
D N
E Y
Alternative Query:
(It's exactly the same method just using UNPIVOT rather than UNION ALL.)
SELECT Account_num,
CASE WHEN COUNT( CASE total WHEN 0 THEN 1 END ) > 1
THEN 'N'
ELSE 'Y'
END AS is_contiguous
FROM (
SELECT Account_num,
SUM( lvl ) OVER ( PARTITION BY Account_Num
ORDER BY CASE lvl WHEN 1 THEN dt - 1 ELSE dt END,
lvl DESC
) AS total
FROM accounts
UNPIVOT ( dt FOR lvl IN ( start_dt AS 1, end_dt AS -1 ) )
)
GROUP BY Account_Num
ORDER BY Account_Num;
WITH cte AS (
SELECT
AccountNumber
,CASE
WHEN
LAG(End_Dt) OVER (PARTITION BY AccountNumber ORDER BY End_Dt) IS NULL THEN 0
WHEN
LAG(End_Dt) OVER (PARTITION BY AccountNumber ORDER BY End_Dt) >= Start_Dt - 1 THEN 0
ELSE 1
END as discontiguous
FROM
#Table
)
SELECT
AccountNumber
,CASE WHEN SUM(discontiguous) > 0 THEN 'discontiguous' ELSE 'contiguous' END
FROM
cte
GROUP BY
AccountNumber;
One of your problems is that your contiguous desired result also includes overlapping date ranges in your example data set. Example A111326 Starts on 3/2/2016 but ends the row before on 3/5/2015 meaning it overlaps by 3 days.
I am trying to figure out how to populate the below NULL values with 1.245 for dates from 07-OCT-14 to 29-SEP-14 then from 26-SEP-14 to 28-JUL-14 it will be 1.447.
This means if the date is less than or equal to the given date then use the value of max effective date which is less than the given date
We could select the last available index_ratio value for given security_alias and effective date <=p.effective_date , so in other words we will need to modify the sql to return from the subquery the index ratio value identified for the maximum available effective date assuming that this effective date is less or equal position effective date
How to populate the value ?
select ab.security_alias,
ab.index_ratio,
ab.effective_date
from securitydbo.security_analytics_fi ab
where ab.security_alias = 123627
order by ab.effective_date desc
Below should be the output
Assuming I understand your requirements correctly, I think the analytic function LAST_VALUE() is what you're after. E.g.:
with sample_data as (select 1 id, 10 val, to_date('01/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, null val, to_date('02/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, null val, to_date('03/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, null val, to_date('04/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, 20 val, to_date('05/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, 21 val, to_date('06/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, null val, to_date('07/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, null val, to_date('08/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, 31 val, to_date('09/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, null val, to_date('10/08/2015', 'dd/mm/yyyy') dt from dual union all
select 1 id, 42 val, to_date('11/08/2015', 'dd/mm/yyyy') dt from dual)
select id,
last_value(val ignore nulls) over (partition by id order by dt) val,
dt
from sample_data
order by id, dt desc;
ID VAL DT
---------- ---------- ----------
1 42 11/08/2015
1 31 10/08/2015
1 31 09/08/2015
1 21 08/08/2015
1 21 07/08/2015
1 21 06/08/2015
1 20 05/08/2015
1 10 04/08/2015
1 10 03/08/2015
1 10 02/08/2015
1 10 01/08/2015