RSUM Function

How do I control where to place a break on the cumulative function. For eg., if I have a drop-down for a region and a quarter in my pivot table, and I have to show the data as cumulative of the weeks within the quarter, how can I do it without the RSUM function summing up beyond the weeks in a particular quarter. How can I make it to break after one quarter and start summing afresh in the next?
I know it can be done by unioning each quarter, but this will not help if its some other dimension with many values.

In your criteria tab go to the formula fot your fact column (FactCol) and use something like the following...
RSUM(FactCol by QuarterCol)

Similar Messages

Problem of rsum function in request

Hi all experts,
I have a problem of rsum function in my request,
i use rsum to get the running sum of count( product number) per month,
but unfortunately, in my query there are some months we didn't have product, so the origin result is like:
Month Product Number RSUM
201001 100 100
201004 100 200
201005 300 500
201007 500 1000
And i have to show all the month even if some months haven't product, like this:
Month Product Number RSUM
201001 100 100
201002 0 100
201003 0 100
201004 100 200
201005 300 500
201006 0 500
201007 500 1000
Anyone can tell me how to get result like this?? Thanks a lot
Xin

Hi saichand,
That's not what i want, i want to get the result like below:
Month RSUM
201001 100
201002 100
201003 100
201004 200
201005 500
201006 500
201007 1000
that means i need to create a request to get the count of product first, and then combine a request with all the month and a static count like 0,
at last to make a rsum function to get the result i want.
But now, the problem is after combine two request, in the abstract request column, it hasn't any formula, so i cant use rsum function to get the result
Thanks,
Xin

Problem: sort and RSUM

Hi,
I have a report in answer, which is combined by 3 single reports and ordered by the date. Also I have a running sum.
With the sort the reports looking similiar to this:
DATE | QTY | RSUM_QTY
01.01.08 | 100 | 100
02.01.08 | 100 | 200
03.01.08 | 100 | 300
03.01.08 | 200 | 300
04.01.08 | 100 | 400
04.01.08 | 200 | 400
04.01.08 | 300 | 400
06.01.08 | 100 | 500
You can see that there's something wrong with the RSUM. Because if it's the same date, the SUM is always the same.
If I have no sort, the report looks like this:
DATE | QTY | RSUM_QTY
03.01.08 | 200 | 200
01.01.08 | 100 | 300
02.01.08 | 100 | 400
03.01.08 | 100 | 500
04.01.08 | 100 | 600
04.01.08 | 200 | 800
06.01.08 | 100 | 900
04.01.08 | 300 | 1200
And that's what I expected ... My funtion for RSUM is added in the result column with the following function: RSUM(saw_2).
So I need the report to be ordered by date, is there anyway to get the RSUM function working correctly?
Thanks
chrissy

I can't duplicate your conditions, so what I suggest may be just a work around. I can think of a couple of thing you can try:
1) Don't sort the column. Instead, keep your report in Table View (if it isn't already) and enable Column Sorting from the Properties of Table View. Then in the Results sort on the Date column.
2) Add a column that would make each row distinct even if the dates and measures total are the same, place it before the date column and hide it. Then apply the sort to the date column.
Tell me if either works for you.

I don;t understand about Rsum

hi all
i don't understand about running functions....
when we have to go for running functions... so far i did not use running functions;
please any one tell me about running functions... i searched about running functions but i am not that much clear on it
Tell me few words about RSUM,..
Thanks

You get more detail in the Help
Just for your information:
This function calculates a running sum based on records encountered so far. The sum for the first row is equal to the numeric expression for the first row. The sum for the second row is calculated by taking the sum of the first two rows of data. The sum for the third row is calculated by taking the sum of the first three rows of data, and so on.
Syntax:
RSUM (n_expression)
Where:
n_expression
Any expression that evaluates to a numerical value.
Example:
The following example shows a query that uses the RSUM function and the query results.
select month, revenue, RSUM(revenue) as RUNNING_SUM from sales_subject_area
MONTH REVENUE RUNNING_SUM
JAN 100.00 100.00
FEB 200.00 300.00
MAR 100.0 400.00
APRIL 100.00 500.00
MAY 300.00 800.00
JUNE 400.00 1200.00
Hope its clear now :)

TODATE alternative in OBIEE 11g (11.1.1.5)

H all,
Is there any alternative to TODATE function? I have got a dashboard with an Year prompt, which allows Multiple Values Selection. I tried using RSUM function in analytics but since user can select multiple years, it does not work.
The solution can be either in RPD or in analytics? Thanks in advance.
Cheers,
Yogi

I have the following fields in my analysis and there is multiple select prompt for YEAR.
Year--Month--Customers--CustomersYTD
Customers is a repository measure --> count(distinct(CUSTORMER_ID).
Cutomer YTD is calculated using TODATE function. I am looking for the alternative to TODATE function.
Thanks in advance.
--Yogi

Running Calculations

Hi All,
Can anybody help me out how to over come the following scenario..
I have created a request which consists of ROW_SEQ, OPERATOR and YTD columns. I need to perform the operation(+/-/*) depends upon OPERATOR column on YTD.
My requirement is like RSUM functionality but need to perform all operations based on OPERATOR column rather than only SUM. How to implement this functionality?
ROW_SEQ OPERATOR YTD GROSS_YTD(I need to include this col)
10 + 10000 10000
20 + 20000 30000
25 - 15000 15000
How to implement GROSS_YTD calculation?
Thanks in Advance,
-Vency

One of the main problems with this kind of a report is that we do not have aggregation functions for / and *. Generally aggregation is defined for functions like average, sum etc. Also, i am not sure whether this would return uniform results for multiple reports. Do you have an uniform order by clause in your report? If the order of the *,- etc varies your entire report output would be wrong. If you only have +, - the order does not matter since they are in the same level in BODMAS. But we are dealing multiple levels of the BODMAS here. One way i can think of is to have a database function and call that function using EVALUATE.
Thanks,
Venkat
http://oraclebizint.wordpress.com

Scaling X - axis in a Line Time series chart

I have a QTD graph built using RSUM function on a measure for year and previous year .
Plotting a X- Y axis plot with both the measures on Y-axis vs calendar days - ( Oct - Nov - Dec ) on X-axis.
Requirement is with out adding any filters to the report can we just display data from Nov -1 to Nov 30 on the Plot. Zoom into the Graph for just the present month data. The reason for doing this is I am using a RSUM function on the column and if I add a filter for present month the QTD values change and will no longer be valid.
Any suggestions appreciated.

Thanks Vinod for the reply.
Second customer doesnt have data on consecutive days but there is no pattern in that.He has data on 3rd,7th,8th,10th,15th day.
But for all these days for the second customer dots are coming but not in the case for First Customer.
Regards

I need some help Regarding Cumulative totals

Hi ,
i have a requirement
1) we have columns calender Periods and actual to date. The requirement is we need cumulate the actual to date. i have applyed RSUM function in column formula.
2) it is working fine. But again the requirement is this cumulative solution is changed based on selected calender periods. but this is not expected.
3) ex:-
calender period actual Units Cumulative Units
oct - 2010 10 10
Nov-2010 9 19
Dec-2010 15 34
Jan-2011 25 59
Feb-2011 12 71
March-2011 19 90
April - 2011 13 103
May - 2011 18 121
June - 2011 24 145
this is the reqiurement but i have a Prompt Like
i hve selected resource categery some xyz and stert period is Jan -2011 and end period is May - 2011
the values disply's like bellow
Resource Category start Period End Period
xyz Jan - 2011 May - 2011
Jan-2011 25 25
Feb-2011 12 37
March-2011 19 58
April - 2011 13 71
May - 2011 18 89
But this is not actual cumulative units the actual value for Jan 2011 cumulative value is 59 not 25
can any one help me regarding this issue
regards,
ramesh

Hi,
Refer this link abt iview properties & set authorization ...
http://help.sap.com/saphelp_nw04/helpdata/en/13/81a66d100011d7b84b00047582c9f7/frameset.htm
Color- you have to change portal theme...
Width- create a new Loyout..
Regards,
Senthil K.

Year to date calculation

Hi experts,
I need to create a request where I calculate year_todate costs for the previous year (one year ago)
Let's say I have the following figures for 2000:
year________Month_________Day______Acutal Coasts
2000________Jan___________1____________10
2000________Jan___________2____________20
2000________Jan___________3____________30
I used the AGO function to retrieve the previous year costs so I end up with the following request:
year________Month_________Day______Acutal Coasts_____Previous year costs
2001________Jan___________1____________5_________10
2001________Jan___________2____________6_________20
2001________Jan___________3____________7_________30
And to calculate the year_todate costs and I tried OBIEE to_date fucntion :TO_DATE(Previous year costs) , the results was:
year________Month_________Day______Acutal Coasts_____Previous year costs
2001________Jan___________1____________5_________10
2001________Jan___________2____________6_________30(10+20)
2001________Jan___________3____________7_________60(10+20+30)
When I look at the results at Month level
year________Month______Acutal Coasts_____Previous year costs
2001________Jan____________5_________100 (10+30+60)
It's summing up all the coasts at day level and I only need the last value of a month, or a year if I want to display the results at year level.
I also tried to use the RSUM function, but because it's only a display function when I filter on a day (for example day 2 JAN) I only got the results without any running sum:
2001________Jan___________2____________6_________20 (because I exculed the other days with the filter)
Any ideas how I achieve this:
year________Month______Acutal Coasts_____Previous year costs
2001________Jan____________5___________60 when I look at the month level
and
2001________Jan___________2____________6_________30 when I look at the day level.
Regards

Thanks venkat for the answer.
It's only solving the problem at month level
year________Month______Acutal Coasts_____Previous year costs
2001________Jan____________5___________60 when I look at the month level
but at day level it's just repeating the value at month level for every day
year________Month_________Day______Acutal Coasts_____Previous year costs
2001________Jan___________1____________5_________60
2001________Jan___________2____________6_________60
2001________Jan___________3____________7_________60
and I want to have the following if I filter on 1JAN and 3JAN
year________Month_________Day______Acutal Coasts_____Previous year costs
*2001________Jan___________2____________6_________30(10+20)*
Regards

Time Series Measure : ToDate

Hi experts,
I have been facing a problem,for which I need your help.
I want to calculate No. of Pending claims till date....
I am trying achieve it using ToDate Function.
ToDate(Cliam Fact.Pending Claim,Year)
But, If I select Year=1999 it will show result like this
Pivot View :
Jan-1999 Feb-1999 Mar-1999 ....................
Pending Claims 10 15 9
ToDate_Pending 10 25 34
I want to calculate pending claims before Jan-1999 also
for ex. till Dec-1998 : if pending claim count is 210
then for Jan-1999 it has to show ToDate_Pending =220..
Can we get this by applying ToDate function on TotalLevel of Period Dimension Hierarchy?
Or do you have any better suggestion for this problem.....
Plz, reply ASAP.

As I understand, it's not a YEAR to date, because you don't want the cumul start from 0 each year. Is it right ?
If so, just try a "running sum" in the pivot table.
You will have these results :
Dec-1998 / Janv-1999 / Fev-1999 / Mar-1999
210 / 220 / 235 / 244
Jan-1999 Feb-1999 Mar-1999 ....................
Pending Claims 10 15 9
ToDate_Pending 10 25 34
If you need to "break" in function of another dimension, then use the RSUM function with "by" (ex : Rsum(measure by Category_product) )
(or Rcount, depending on what is your aggregation of your measure).

How to use groupby function in OBIEE ANSWERS ???

Hi ,
I need to find TOP 1000 client names on each product and I need to group the data by product. I am unable to group the data in Answers , I dont have access to the repository.
I used the Top function but unable to group the data. Please help me...

Why didn't using TOP work for you? What kind of results did you get?
I can also suggest using custom ColumnCount with RSUM () in it, applying filter RSUM()<=1000, and sorting by product amount DESC
and definitely your column order should be something like this - product group, client, product_amt, rowcount (you can actually remove this once you create filter for it)

Running Aggregate Functions, Attribute or Measure?

Hi Guys:
I wonder if this is a bug in BIEE 11g? I am using 11.1.1.6.
My requirement is quite simple, I have a measure, and I want to show this measure as a cumulative curve by month. For example, let's suppose there are three columns in Criteria:
Month, Amount, RSUM(Amount)
Then I plot a curve by RSUM(Amount) vs. Month. In BIEE 10, there is no problems. However in 11.1.1.6, I got an error. I look into the table and found that, after using running aggregate functions, like RSUM or MSUM, the column becames Attribute, instead of Measure. In other word, although I can see the RSUM(Amount) column in Criteria page shows a yellow icon, it turns into blue in Results page! Since it becames Attributes in Results page, I cannot drag RSUM(Amount) into Measures area in Line Graph. I can see the correct values in Table view, but got error in Line Graph view.
Why the column properties are different between Criteria and Results? And are there any workaround to plot a cumulative curve? I dont want Parete graph.
Thanks
Larry

Hi Larry,
I do not think this is any bug in .6 and I just gave a try successfully to confirm that too. My test case follows
1. Created an analysis with Period.Month and Measures.Value
2. Added another measure and changed the formula of Measures.Value as RSUM(Measures.Value) and still see it to be a measure as expected.
3. Navigated to results section and added all kind of graphs to notice that it is behaving as expected.
Can you let us know, what error are you hitting at in the graph?
Thank you,
Dhar

RSUM and level based measure

Hi all,
Is it possible to use the RSUM display function with level based measure. RSUM(LBM) in Answers.
I tried to do so but the RSUM is never correct,so i'm wondering if I'm doing something wrong or the RSUM is not working correctly with level based measures.
Regards

What is the month level key column functionality in your time dimension?
it should be in format of yyyymm but not in just mm ..

OBIEE RSUM() with partition by

Hi All,
Can anyone suggest me that how to get the below result in OBIEE:
Row Number Title Date
1 ABC 1/1/2012
1 ABC 1/2/2012
2 DEF 1/1/2012
3 GHI 4/1/2012
Currently I am using RSUM() and only able to get the below result which is not my expected result:
Row Number Title Date
1 ABC 1/1/2012
2 ABC 1/2/2012
3 DEF 1/1/2012
4 GHI 4/1/2012
I want to get the running distinct number of title instead of getting the running number of all records.
Any help would be appreciated. Many thanks.
Thanks & Regards,
Benard

Hi SVee,
Thanks for the reply. I have tried your suggestion but it is only partially working.
I have use suppress and use count(title) function and it shows below result:
Row Number Title Date
1 ABC 1/1/2012
ABC 1/2/2012
1 DEF 1/1/2012
1 GHI 4/1/2012
If I add in RCOUNT(title) function, it will still shows:
Row Number Title Date
1 ABC 1/1/2012
2 ABC 1/2/2012
3 DEF 1/1/2012
4 GHI 4/1/2012
I even set suppress to Title column as well, but it just didn't work well.
Thanks & Regards,
Benard

Aggregate Functions and Difference

I had posted this question earlierRCOUNT and RSUM but if that was too many details, let me simplify it here.
Month_Number -----Count(Orders)----------Increase In Sales
1-------------------------100------------------------- -
2-------------------------400------------------------- 300
3-------------------------280------------------------- -20
This is what Im looking for. As I have to use the Count() I am using Group By on Month_Number.
RSUM didn't work in this case. Any solution to get this working ?
Thanks
Edited by: user558238 on Jan 19, 2009 11:06 PM

As you have only one period in the difference.
For instance :
Month 2 - Month 1
and not
Month 3 - Month 1
You can use this formula :
MSUM (CountOrders, 2) - 2 * CountOrdersIt calculate a moving sum on 2 periods and subtract 2 sum of the current period.
Otherwise, you can use the Time function series (AGO, TODATE) but it's only possible in the repository layer and not in the Business presentation layer :
http://gerardnico.com/wiki/dat/obiee/bi_server/design/dimension/time_dimension_parameters_for_ago_and_td_functions
AGO(CountOrders, SH.TimesDim."Month", 1)An other solution will be the use of a sql function.
Success
Nico

RSUM Function

Similar Messages

Maybe you are looking for