3 weeks Average

Hi all,
I have an issue regarding the calculation of average sales of the last three weeks. So, the average sales is:
week 3 --> (week 1 + week 2 + week 3) / 3
week 4 --> (week 2 + week 3 + week 4) / 3
week 5 --> (week 3 + week 4 + week 5) / 3
week 6 --> (week 4 + week 5 + week 6) / 3
(etc.)
So, the report needs to look like this:
week sales   average sales
1     10
2     15
3     10      11,7 = (10 + 15 + 10) / 3
4     20      15    = (15 + 10 + 20) / 3
5     25      18,3 = (10 + 20 + 25) / 3
6     30      25    = (20 + 25 + 30) / 3
A requirement is also that all the weeks need to be displayed. This rules out the possibility to start with a pop-up, where the user enters a week and then use a restricted key-figure with an off-set.
I have also considered to make restricted keyfigures for each week, as also suggested in /forums/viewtopic.php?t=84733. However, the problem with this, is that a year does not have a fixed number of weeks. This can be either 52 or 53.
Thanks,
GAL

Hi,
You need to make use of Offsets.
Create offset for Week and then apply sales figures.
Regards
Govind.

Similar Messages

Calculate Four Week Average SellThru In RPD

Hello All,
We have a requirement to calculate the four week average sellthru dynamically in OBIEE. Basically we need to expose SellThru Average as a column in presentation layer. The calculation of this column should use the below logic. Also see below for data model details too.
1. The SellThru Average is dependent on another metric called SellThru
2. The data in fact table is at a grain of Item and Customer. These two are confirmed dimensions.
3. Rules to calculate the SellThru Average
a. If the customer is looking at the Week10 data at item level, the average should be calculated using Week5, Week6, Week7 and Week8 SellThru.
b. If the SellThru is available for all the above weeks, then the formula for average will be (Week5 SellThru+Week6 SellThru+Week7 SellThru+Week8 SellThru)/4
c. If the SellThru is not available for Week5, then we need to go back and see if SellThru data is available with in last 26 weeks for that Item. If yes then the denominator will be 4 and numerator stays the same
d. If the SellThru is not available with in last 26 weeks, and only available for Week6, Week7, Week8 then the formula would be (Week6 SellThru+Week7 SellThru+Week8 SellThru)/3
e. If the SellThru is not available with in last 26 weeks, and only available for Week7, Week8 then the formula would be (Week7 SellThru+Week8 SellThru)/2
f. If the SellThru is not available with in last 26 weeks, and only available for Week8 then the formula would be (Week8 SellThru)
g. If the SellThru is not available with in last 26 weeks, and only available for Week5 then the formula would be (Week5 SellThru+Week6 SellThru+Week7 SellThru+Week8 SellThru)/4.
In all the above cases, for whichever week the SellThru is not available, substitute with 0
These rules needs to be applied at all the higher levels for both Item and Customer dimensions. For example, if the user is looking at Segment level which is the next level to item in Product hierarchy we need to check if the SellThru is available for any of the item under that segment for the respective weeks and apply one of the above formula accordingly.
We were able to achieve this using OBIEE logical columns and repository variable for Current Week alone. But we are stuck at calculating the averages for previous weeks. Meaning, if the user is viewing the data for Previous week, we need to go back from that week. It's basically dynamic in nature. User can also look at two or more weeks data at the same time and the average needs to be displayed as per that.
I hope i have explained the requirement OK. Any help in this will be well appreciated. Thanks in advance.
Siva

Hi Thomas,
Create a selection in which you will key figure KF3 and a calweek with restriction with fromcalweek and to calweek variables.
The tocalweek is a formula variable which is derived from replacement path of characteristic calweek.
and fromcalweek is offset of tocalweek variable - 6.
So for each week it put toweek as that weekno and fromweek as weekno - 6 and brings the KF3 value in that range.
Regards
vamsi

Calculating weekly average in reports

I have a report that looks like:
Vol previous 7 day average
12
34
45
56
34
98
12 41.6
45 46.2
76 52.3
89 etc...
12
What would be the best way to do the previous 7 day average values in Reports 3.0.
Thanks for any assistance.
null

I have a table like this:
ViewID
ViewDateTime
I would like to have the weekly average of views and
have an output that looks like:
[Weekly Avg] [Week Starting]
Any help would be greatly appriciated.If I have well understood, then this will help you: (format is the format of your choice)
SELECT Avg(ViewID) "weekly Avg",
TO_CHAR(TRUNC(MIN(ViewDateTime), 'WW') + 1, 'format')
FROM t
GROUP BY TO_CHAR(ViewDateTime, 'YYYY') || TO_CHAR(ViewDateTime, 'WW')I began editing before your explanation so you can suppress the year if all your records are in the same one.
Message was edited by:
Michel

Query to find out weekly average of job execution time

I have a table which has the following details:
CREATE TABLE JOB_DETAILS (JOB_START_DATE DATE,
JOB_END_DATE DATE,
JOB_START_TIME NUMBER,
JOB_END_TIME NUMBER,
HOURS_OF_EXECUTION NUMBER,
DETAILS VARCHAR2(20));
INSERT INTO JOB_DETAILS VALUES ('01-FEB-2007','01-FEB-2007',10.04,20.09,10.04,'WEEKDAY');
INSERT INTO JOB_DETAILS VALUES ('07-FEB-2007','07-FEB-2007',00.00,00.00,00.00,'WEEKEND');
commit;
Job_Start_Date Job_End_Date Job_Start_Time Job_End_Time Hours_Of_Execution DETAILS
1/2/2007 1/2/2007 10:04 20:09 10:04
1/7/2007 1/7/2007 0:00 0:00 0:00 Weekend
Our jobs wont run on week ends and on holidays where we see "Hours_of_execution" as 0.
Week means Monday to Friday for me.
(1) I want a query which gives me a weekly,monthly average of how many hours the job ran.
(2) I want to find out the week, when the average execution of the job is > 10 hours.
can anyone help in framing this query to get the details?
Thanks

Why make it so hard for yourself by removing the time from the date and storing it separately?
alter session set nls_date_format='DD/MM/YYYY HH24:MI:SS'
session altered
WITH t AS (SELECT SYSDATE start_date, SYSDATE+0.61 end_date FROM dual)
SELECT start_date, end_date, numtodsinterval(end_date-start_date,'DAY') execution_time
FROM t
START_DATE END_DATE EXECUTION_TIME
11/12/2008 11:22:21 12/12/2008 02:00:45 +00 14:38:24.000000

Average quantity for Prior weeks

Hello Friends,
I have a requirement to calculate prior 4 weeks quantity average and price average.
And also I have to calculate price variance by substracting current weeks price with previous weeks price.
I have price and quantity are in KF, and calyear/week in free characteristcis.
Can you please help me?
Thanks
Priya

Hi,
I have followed below suggestions and it didn't give me any results.
Currently, I have 3 KF, quantity, price and total price.and Calweek/year which is restricted to current calweek and previous week in KF section.
And it displays results like this, and which is what I wanted.
07.2007 06.2007
Quantity Price Total Price Quantity Price Total Price
In addtion to this results, I have to display prior 4 weeks average of Quantity,Price and total price.
These are the steps I have followed:
Created a structure, 4 new selections for four prior weeks, with offsets on calyear/week till -4, by having quantity in the selection. I am not able to do the formula as price and quantity doesn't appear for those four columns I have created for prior weeks.
I appreciate your help .
Thanks
Priya.

Group and Sum/Average

Hello.
I have an excel structure here, this data is extracted from our company's software. The record was recorded from Dec 5, 2013 to Dec 11, 2013
Job Operator
Added
Job Opened
Job Revenue Recognition Date
Shipment ID
Joebeth Cañete
05-Dec-13
19-Dec-13
IMM
S00038408
Joebeth Cañete
05-Dec-13
19-Dec-13
IMM
S00038412
Joebeth Cañete
05-Dec-13
19-Dec-13
IMM
S00038414
Joebeth Cañete
05-Dec-13
10-Dec-13
IMM
S00038440
Sharlah Faye Solomon
09-Dec-13
09-Dec-13
IMM
S00038501
Rachel Rosales
09-Dec-13
09-Dec-13
IMM
S00038502
Sharlah Faye Solomon
09-Dec-13
09-Dec-13
IMM
S00038503
Sharlah Faye Solomon
09-Dec-13
09-Dec-13
IMM
S00038504
Sharlah Faye Solomon
09-Dec-13
09-Dec-13
IMM
S00038506
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038508
May Jane Herrera
09-Dec-13
09-Dec-13
IMM
S00038509
Joebeth Cañete
09-Dec-13
17-Dec-13
IMM
S00038510
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038512
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038513
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038514
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038515
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038516
Sheena Dagandan
09-Dec-13
09-Dec-13
IMM
S00038518
Joebeth Cañete
10-Dec-13
10-Dec-13
IMM
S00038523
Sharlah Faye Solomon
10-Dec-13
10-Dec-13
IMM
S00038524
May Jane Herrera
10-Dec-13
10-Dec-13
IMM
S00038525
May Jane Herrera
10-Dec-13
10-Dec-13
IMM
S00038526
Rachel Rosales
10-Dec-13
10-Dec-13
IMM
S00038528
May Jane Herrera
10-Dec-13
10-Dec-13
IMM
S00038530
May Jane Herrera
10-Dec-13
10-Dec-13
IMM
S00038531
May Jane Herrera
10-Dec-13
10-Dec-13
IMM
S00038532
Joebeth Cañete
10-Dec-13
10-Dec-13
IMM
S00038533
Rachel Rosales
10-Dec-13
10-Dec-13
IMM
S00038534
Sharlah Faye Solomon
11-Dec-13
11-Dec-13
IMM
S00038541
Sharlah Faye Solomon
11-Dec-13
11-Dec-13
IMM
S00038542
Sharlah Faye Solomon
11-Dec-13
11-Dec-13
IMM
S00038543
Sharlah Faye Solomon
11-Dec-13
11-Dec-13
IMM
S00038544
May Jane Herrera
11-Dec-13
11-Dec-13
IMM
S00038548
May Jane Herrera
11-Dec-13
11-Dec-13
IMM
S00038549
Rachel Rosales
11-Dec-13
11-Dec-13
IMM
S00038554
Rachel Rosales
11-Dec-13
11-Dec-13
IMM
S00038555
Rachel Rosales
11-Dec-13
11-Dec-13
IMM
S00038557
Rachel Rosales
11-Dec-13
11-Dec-13
IMM
S00038558
Now, my boss wants to have a matrix like this.
Job Operator Added Job Opened Shipment ID
Employee A
Total for (n) weeks [of Employee A]
Average Shipment Daily:
Average Shipment Weekly:
Employee B
Total for (n) weeks [of Employee B]
Average Shipment Daily:
Average Shipment Weekly:
Inshort, he wants to group the records by Job Operator and get the average number of shipments daily and weekly. Is it possible to do this job?

What' the define of the 'Week' you mentioned?
It would be benefit for us to find the solution if you can provide the output you want.
E.g..
Job Operator
Total for (n) weeks
Average Shipment Daily
Average Shipment Weekly
Joebeth Cañete
2
1
3.5
But in any case, you can try to create a pivot table to analyze the data.

Calculating 13 period average

I need to calculate 13 period average valuesi.elets say if its feb,fy06 then it should calculate theaverage value of jan to dec of fy05 plusjan,fy06.i'm trying out something like this IF @ISMBR(feb) and @ISMBR(fy06)xyz=@SUMRANGE("member1",@CURRMBRRANGE("TimePeriods",LEV,0,-1,10))"member1"->FY06->P1)/13); I'm not sure how to go back toprevious yr. and calculate the average value.I would appreciate if somebody can help me outwith this.

Whenever we define fiscal year, we make certain calculations easier and others more difficult. Making fiscal year its own dimension outside the time dimension makes year-to-year comparisions easy, but calculations like the one you propose a bit more difficult. I keep fiscal year within my time dimension, so the @prior function makes calculations like you propose fairly easy. With a separate fiscal year function, you'll need to invoke @prior on the fiscal year, along with using @lsiblings to identify prior periods in the current fiscal year. That, @sumrange and a few condtionals and you'll have a formula I regularly use a separate, low-overhead dimension to contain my time series calculations. For purpose of discussion I call the dimension "horizon" as that was the name used by the consultant I learned it from years ago. Horizon has one member where data is stored, "period", and includes all of the calculations for things like 13-week average, prior year, etc., all as dynamic calcs (2pass as necessary). There are some advantages to doing YTD and related time functions here as well. By doing this, I avoid duplicating formulas, moving UDA or otherwise modifying the outline. This horizon dimension really becomes a real benefit when used with Analyzer or other reporting tool in conjuction with substitution variables that track current period and other key periods. It minimizes manual maintenance, while making all of the time series related values available for any periiod, not just the current period.

Working out an average

SAP Business Objects Infoview
Hi one and all,
I need to create a measure variable for a report so that I can calculate the average for the past 9 month - excluding the first quarter (April May and June). #
                             Apr   May       Jun        Jul    Aug   Sep     Oct    Nov     Dec    Jan     Feb   Mar   YTD
Patients             18660 18293   19021 19608 15952 18083 19072 16874 16228 17088 16928 19034 214,841
(example)Avergae 15,600 15,600 etc etc etc    15,600 15,600 15,600, 15,600 etc etc etc
So I have the table as above and I would like to have a row above this with an average from July through to March and so as to have a means to compare the data.
The measure needs to be a calculation as the report drillable so the average needs to change according what is being drilled.
On top of this monthly average I need to do the same thing but have a weekly average, again with the same criteria as above for the month.
Any help much appreciated!!!
Many Thanks in advance,
Nick

Many Thanks Tanveer, you are a star!
Much appreciated!
If I could ask one more thing, if I wanted to do a monthly variance between the two montlhy figures Patient and Average, what would the variable be for that? Also a variance percentage variable?
Thanks again,
Nick
Also I would like to add the avergae line to an SPC chart that I currently have but cannot change the formula as I have done for the cross tab, abyway of brining this Average figure into the SPC chart at all?
Many Thanks,
Nick

Not showing all data in a grid - a summary

Hello all,
I have a very simple spreadsheet (created from a template on an ipad and transferred back and forth to a Mac Book Pro) which tracks my weight loss, body fat etc.
Each day (probably far too often but I'm eager at the start of a diet) I record the figures to a new row so row 1 are the figures for my oldest date, row 2 is the next day etc and this has been going on for almost 4 weeks.
I show a simple auto created chart so I can track my weight loss. The date is shown along the bottom X axis so it shows a tick mark and date for every day. The Y axis for one chart shiows my weight loss and for another chart it shows my body fat loss.
Is there any way I can make the charts show tick marks AND VALUE POINTS (say) once per week or every 2 days? Or perhaps eventually once per month. I know I can change the tick mark intervals shown but my values and points are still shown on a daily basis (the same as they have been recorded in the spreadsheet). With one value per day, my chart is gradually growing so it will no longer fit the screen or worse still no longer print onto A4.
I realise I could record my data less often or delete rows from my spreadsheet so it has one row per week but for the moment at least I'd like to stick to recording my weight on a daily basis but maybe chart it on a weekly or monthly basis. I don't mind having a completely different chart showing weekly weight loss but I can't see a chart or table option anywhere that says only show data on a weekly (7 row basis). Perhaps I've totally missed the point of how this should be done. For example, should I have another table that some how calculates a weekly average from the first table and then I show a chart for that second table. If so, I think my current knowledge is not sufficient to achieve that!
Thanks
John.

Hi John,
Thank you for your feedback. I used a 5 day average, forgetting that you asked for a weekly average. But I am sure that you can change the formula.
I don't seem to be able to get connecting lines between the points on the chart.
Yes, I had trouble getting connecting lines in Chart 2. I don't know the answer to this. Let's hope that the High Priests in this forum can help.
I wrote:
If you miss weighing on one day, the weight will be blank. Revise the AVERAGE formula to exclude zeroes.
Again, let's hope that the High Priests in this forum can help.
Over to the Gurus!
Regards,
Ian.

Viewing past days in the Health app?

I like to check how many steps I take each day, but if I forget to check before midnight, I lose that data. I can't figure a way to go back and read previous day's step totals. It's so frustrating. I can see weekly averages, which I don't care about. Why isn't there a simple calendar view where you can select any day and see that day's total information? Does that just make too much sense? So if I forget to check on a given day, that information is gone forever.

If you set up renditions properly, only one rendition should appear in the library. In the set of renditions, make sure that the Folio Name, Publication Date, and Product ID values are identical. In a public Newsstand app, publish the folios around the same time, and when they're done, send the push notification.

Hope to have crosstab result

I have following SQL to get weekly average,
SELECT next_day(dt-1,'FRIDAY') FriDate,avg(midpoint),commodity_id,location_id
FROM price where location_id in(50,22,33,48)
and commodity_id in(7,7,12,8)
and dt>=to_date('4/12/1996','mm/dd/yyyy')
and dt<=to_date('12/31/2000','mm/dd/yyyy')
group by next_day(dt-1,'FRIDAY'),commodity_id,location_id
result like this:
FRIDATE AVG(MIDPOINT) COMMODITY_ID LOCATION_ID
4/12/96 222 7 22
4/12/96 212 7 33
4/12/96 217 7 50
4/12/96 495 12 22
4/12/96 490 12 33
4/12/96 485 12 50
4/19/96 222 7 22
4/19/96 211.2 7 33
4/19/96 217 7 50
4/19/96 481 12 22
4/19/96 474 12 33
I wold like to have crosstab report as following
FRIDATE AVG(MIDPOINT) of COMM_ID(1)LOC_ID(1) AVG(MIDPOINT) of Comm_ID(2)Loc_ID(2)....dynamic showing both ID on field title
4/12/96 222(7,22) 212(7,33).....show average(not the ID)
4/19/96 222(7,22) 211.2(7,33)
How I change the query to achieve this.
Thanks

The processor can be upgraded; it is socketed. However, virtualization must be implemented in the motherboard and BIOS and without that, upgrading the processor will do no good. Which processor do you have?
Here is the Service Manual:
Manual
See page 120....virtualization is possible on some models but I am going to assume only the models with a Core2 Duo processor. Celeron and Pentium Dual models use a different chipset on the motherboard so even if you put in a Virtualization capable CPU it would not work.
If this is "the Answer" please click "Accept as Solution" to help others find it.

Report Filtering

Hi,
I am using CR 2008. I have a crosstab where I need to show sales for Current & Prior year for Current Week(CW) through 4 previous weeks. I also need a Rolling average of Sales for 8 Previous Weeks from CW. I need to place this field right next to the Crosstab in the same Group.
Ideally I want to have just one query and get all 8 weeksu2019 data back. I can use record select expert to restrict the data shown in crosstab to only 4 weeks. But while calculating the Rolling 8 week average I am not able to ignore this filter. Is there a way to apply the filter specific to the Cross tab or someway to ignore the record selection criteria for the Rolling 8 week Avg field. (Like the Nofilter() function in WebI.)
Thanks in advance.

you can build a subreport for the 8 week average which is handled as a seperate report.
it will be tough placing anything next to a cross tab due to the way it previews.
you may want to create a regular report and border it.

Give me some suggestion in this case.

In my application. My database table have one year data about how many hours I watch TV in random day( which means I only count the hour and date when I watched TV and store the hours and date in database). I wanna get the data of the average hours each week during the whole year.
So, now I have data of certain date(eg: 20/04/2007) and hours(3 hours watching TV), and I have one year such kind of data. How can I get each week average hours I watched TV during one year?? Give me some algorithm or suggestion please, thanks.
Message was edited by:
Mellon

Ok, I make the question clear. For example, if I assume each year have 48 weeks. In my database, I have the date,hour pair data, only the date which I watched TV will store in database. What I want is for each week, sum the hours in each day which will get the total hours of that week, than divided by 7 which will get the average hour in that week.
Now, I want to get 48 such kind of average values which represent the weekly average hour in the year.
Again, the database only hold the data of the date, and the hour refer to the date. If I did not watch TV , there will no data in database for that day.
My question is how to get those 48 weekly average values according to the database structure. Thanks.
Message was edited by:
Mellon
Message was edited by:
Mellon
Message was edited by:
Mellon
null

Has apple stopped Raw compatibility updates??

I Just looked at all of the raw updates over the past year. Shortest time between updates.... 3 weeks, longest time 9 weeks, average, 6.3 weeks. Time since last update 10 weeks.
i Have brought this up in other forums and people are saying either Apple is dropping support for iphoto and Aperture and to not expect any further updates. Others say dont worry Apple is committed to photographers.
THe current evidence is supporting the naysayers. If they do not update soon.... I can read the handwriting on the wall and its off to LR... but I really hope not.

The subtext in your question is really all about Apple's priorities: how much energy is it putting into meeting the needs of pro and amateur photographers who use cameras that shoot RAW?
I doubt that Apple is too busy to find the time (although they've got a lot of money to count), and even if that were so, it says something about priorities. But there are tons of alternatives; even open source RAW conversion applications like dcraw/RAWTherapee have a more recent list of supported cameras.
But consider RAW support and what it means. Traditionally Apple doesn't try to support tons of different kinds of hardware. If you want your hardware to work with Apple's system it's sorta on you to make it happen. Consider another image input: scanners. Canon has to provide the driver; Apple now bundles those in system upgrades, but I doubt they have someone writing them. It's up to Canon to do that. Most camera manufacturers don't publish specs about their RAW files; much of the work has been done by sorta reverse engineering them. dcraw started doing that a long time ago, and is behind some of the software you see out there today (RAWTherapee is an example). That's not necessarily bad, I guess, since the intial processing of RAW files gives developers reason to say their software (CaptureOne, LR, DxO, etc) does a better job of this than others.
And Apple has more to do than say PhaseOne. Their camera RAW updates also work with the image i/o framework to identify the filetype and make it available to more than Aperture or iPhoto; that's why you can "see" the RAWs in icon views in Finder windows, in QuickLook, and in other applications that rely on Apple for RAW support. So I imagine it's not as easy for them as for one-off application developers. I'd also be curious to know whether the camera manufacturers support Apple's work. Do they give them the specs for the RAWs? There's probably no reason why, in addition, that the manufacturers couldn't make a qlgenerator plugin (for QuickLook) that would make their RAW files visible in the Finder. I have one made by EyeTV, for example.
So while we're all speculating here, I'd say that it's possible that some of this might be down to your camera manufacturer. Love to know if that's true. And maybe the whole system for doing RAW is changing. Perhaps Apple is putting more onus on the camera guys to do the work; can't say I'd blame 'em.
I don't know if it's still true, but some folks have had to use Pacifist to load the RAW support if they didn't have iPhoto or Aperture installed. This led someone to write a qlgenerator for folks with that dilemma (http://rene.rebe.de/2013-08-11/exactimage-quicklook-w-camera-raw-support/). Although it wouldn't do the conversion for you, it might make your currently unsupported RAW files more usuable on your Mac.
While I think Apple whiffed with the whole Aperture thing, and I can see why some are very annoyed with them, it's not like there's a lack of alternatives.

Need help with Monthly Transaction By Hour Report

Post Author: cbcombs
CA Forum: General
Hello,Newbie needs to write a transaction report that is grouped by days of the week for a calendar month. It will have one column for each day of the week and each of those columns will contain a whole months of week day data. Then it will be grouped by each hour of the day. It will contain the number of transactions for each hour. And then a weekly average of the number of transactions for each hour of the day. The week starts on a Monday and goes through Sunday. They are probably going to want the dollar amount of the transactions too. Below is an example with the first two rows of data populated. Monthly Transaction Report For August 2007 Mon Tue Wed Thu Fri Sat Sun Average08:00 0 2 2 1 10 5 2 3.1409:00 25 3 5 4 5 8 6 4.43 10:00 5011:00 10012:00 15013:00 15013:00 12514:00 10015:00 5016:00 1017:00 2 Any help would be appreciated.

Post Author: kbrinton
CA Forum: General
You could use the cross tab expert and set the columns to the day and the row to hours with it summing your transactions. So for the rows choose your date field and then choose Group options and have it set to "for each hour". Then the column choose your date field and have it set to "for each day". Then have it summarize your transactions.
Hope this helps!!

3 weeks Average

Similar Messages

Maybe you are looking for