TM1 Data change error

[Totti.Ning]

Dear all.

I am new comer for TM1 product. now find a very odd issue, that the data of grid will have little change when I changed the dimension member from one to another, in fact that the second data quote from the first data . ???

wish get your kindly response. thx.

[Gregor Koch]

Hi
Can't wait for the answers on this.
Good luck.

[Alan Kirk]

Dear all.

I am new comer for TM1 product. now find a very odd issue, that the data of grid will have little change when I changed the dimension member from one to another, in fact that the second data quote from the first data . ???

wish get your kindly response. thx.

First, welcome to the forum.
Second, please read the Request For Assistance Guidelines.
Third, after reading the guidelines you may want to upload some small screenshots using the "Upload Attachment" option at the bottom of the posting window. I think that it would help people to understand your issue more clearly.

[Alan Kirk]

Hi
Can't wait for the answers on this.
Good luck.

First timers should be allowed a little slack. It's the repeat offenders who sometimes seem to crave the appearance of sharpened tongues.

[Gregor Koch]

Hey, I didn't ssssay anysing wiss a ssssarpened tongue. I had the exact same issue before and never got it resolved.

With a promise to try and help, welcome to the forum, Totti.

[Alan Kirk]

Hey, I didn't ssssay anysing wiss a ssssarpened tongue.

I've always thought of this place more as Gryffindor than Slytherin. The latter is more Big Blue's bag.

I had the exact same issue before and never got it resolved.

The problem is that I'm not sure how you can tell what the issue actually is from that.

Leaving aside the basics of which 9.5 version this is (9.5.0, 9.5.1 or 9.5.2):

now find a very odd issue, that the data of grid

Which grid of data are we talking about? Cube Viewer? A slice? If so, an Active Form or a standard slice? Perspectives? Architect?

will have little change

What's the working value of "little change"? Some numbers changed? No numbers changed? Was recalculation forced by [F9]?

when I changed the dimension member from one to another

Which dimension member? A title dimension? A row dimension? A column dimension? Is auto recalculation on (assuming that it's really Cube Viewer)? Because if it isn't there's no way that the numbers can't change; in fact they'll vanish until a recalculation is done.

, in fact that the second data quote from the first data . ???

So one set of elements returns the same set of numbers as a different set? Are there consolidations involved? Are there rules involved? Is there a slice in Excel with manual calculation mode involved?

I agree that we're always ready to try to help, but this one really has to go into the "much, much more information required" bin for the present time.

[Gregor Koch]

Alan, now that you mention all these questions I think my issue was of a different nature.

[Totti.Ning]

Thanks for so more reponses for me. sorry for offered less information, now I try to upload two pictures, wish that can help you know my issue,
in fact all data of pic MR2+10 come from by rules

[{'MR_1+11','MR_2+10', 'MR_3+9','MR_4+8','MR_5+7','MR_6+6','MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Jan']=N:['Actual','Jan'];
[{'MR_2+10', 'MR_3+9','MR_4+8','MR_5+7','MR_6+6','MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Feb']=N:['Actual','Feb'];
[{'MR_3+9','MR_4+8','MR_5+7','MR_6+6','MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Mar']=N:['Actual','Mar'];
[{'MR_4+8','MR_5+7','MR_6+6','MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Apr']=N:['Actual','Apr'];
[{'MR_5+7','MR_6+6','MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'May']=N:['Actual','May'];
[{'MR_6+6','MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Jun']=N:['Actual','Jun'];
[{'MR_7+5','MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Jul']=N:['Actual','Jul'];
[{'MR_8+4','MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Aug']=N:['Actual','Aug'];
[{'MR_9+3','MR_10+2','MR_11+1','MR_12+0'},'Sep']=N:['Actual','Sep'];
[{'MR_10+2','MR_11+1','MR_12+0'},'Oct']=N:['Actual','Oct'];
[{'MR_11+1','MR_12+0'},'Nov']=N:['Actual','Nov'];
['MR_12+0','Dec']=N:['Actual','Dec'];


however you can see some data have change at MR2+10 pic when I change the dimension to MR2+10.

[Martin Ryan]

I take it you didn't build this cube and you're trying to work out what it does? It looks to me like it's a rolling forecast cube. It's a bit hard to tell with those element names.

Actuals are actuals. They will be loaded by some process or other from your general ledger or other source system.

I think MR_2+10 is the rolling forecast as at February. So the year to date months (i.e. January and February) are retrieved from the actuals. However all subsequent months are left open for the forecast to be entered.

So by choosing "Actual" or "MR_2+10" you are choosing different scenarios. The first is actual data. The second is the forecast as at close of February.

Even if I'm not quite right about the intention of the cube, that should at least explain why your data moves when you change the scenario. By the way, I would expect the values in a cube to always move when I change one element from another. E.g. changing the year from 2011 to 2012 will change the numbers. Changing "External" to "Internal" should change the numbers.

HTH,
Martin

[Totti.Ning]

Hello Martin

thanks for your so nice interpret, yeap, below two snapshts are from cube A that all data come from cube B by rules, B also have same dimension with Actual and MR2+10…… members, for cube B, all actual data come from bussiness data by input by me( for testing ), then I change the member to MR2+10 and copy Mar~Dec data then paste to MR2+10's Mar to Dec. thus the data for Actual and MR2+10 are same for cube B, in fact the below Actual snapshot have the same value with cube B, however that changed in MR2+10 ...

[asutcliffe]

Hello Martin

thanks for your so nice interpret, yeap, below two snapshts are from cube A that all data come from cube B by rules, B also have same dimension with Actual and MR2+10…… members, for cube B, all actual data come from bussiness data by input by me( for testing ), then I change the member to MR2+10 and copy Mar~Dec data then paste to MR2+10's Mar to Dec. thus the data for Actual and MR2+10 are same for cube B, in fact the below Actual snapshot have the same value with cube B, however that changed in MR2+10 ...

Hi,

What I think you're saying is that you manually input the 'Actuals' into cube B and then manually copy the figures in 'Actuals' for 'Mar' to 'Dec' into MR_2+10 in cube B. You say that the data in cube A is pulled from cube B via a rule. Is that correct?

However the only rules you've posted are internal to cube A - that is don't show how the data is pulled from cube B. Plus, as Martin has pointed out, the rules you've shown us only populate 'Jan' and 'Feb' for the element 'MR_2+10' and your screenshot shows that the numbers are the same for these months.

So if your problem is that the values for 'Mar' to 'Dec' are not the same for 'Actuals' and 'MR_2+10' you need to look at how these values are being populated. Learning how to use the rules tracer can be pretty useful in these situations:

http://publib.boulder.ibm.com/infocente ... F0007.html

I'm guessing though, that Martin is right and cube A is doing some sort of rolling forecast. In which case you wouldn't necessarily expect 'Actuals' and 'MR_2+10' to be the same.

If you're stuck, perhaps you could post more of the rules in cube A?

PS - call me old fashioned but I wouldn't expect to see anything in 'Actuals' for the second half of 2012.

[Totti.Ning]

['Net-Turnover']=N:DB('CT_P&L',!Internal_External Budget,!organization,!Versions,'IAS €',!Years,!Months,'Net-Turnover');
['Operating Income']=N:DB('CT_P&L',!Internal_External Budget,!organization,!Versions,'IAS €',!Years,!Months,'Operating Income IAS');
['Steuersatz HBI']=DB('CT_Premises',!Internal_External Budget,!organization,!Versions,!Years,!Months,'Steuersatz HBI %');
['Steuersatz IAS']=DB('CT_Premises',!Internal_External Budget,!organization,!Versions,!Years,!Months,'Steuersatz IAS %');
['- Steuererstattung/+ -last']=N:['Operating Income']*['Steuersatz IAS'];


pull data rules from Cube B to A ... there is no feeder in this rules

[asutcliffe]

['Net-Turnover']=N:DB('CT_P&L',!Internal_External Budget,!organization,!Versions,'IAS €',!Years,!Months,'Net-Turnover');
['Operating Income']=N:DB('CT_P&L',!Internal_External Budget,!organization,!Versions,'IAS €',!Years,!Months,'Operating Income IAS');
['Steuersatz HBI']=DB('CT_Premises',!Internal_External Budget,!organization,!Versions,!Years,!Months,'Steuersatz HBI %');
['Steuersatz IAS']=DB('CT_Premises',!Internal_External Budget,!organization,!Versions,!Years,!Months,'Steuersatz IAS %');
['- Steuererstattung/+ -last']=N:['Operating Income']*['Steuersatz IAS'];


pull data rules from Cube B to A ... there is no feeder in this rules

Is cube A called 'CT_P&L' or 'CT_Premises'? If not you're pulling data into cube A from not one but two different cubes which might explain some things.

I think to get anywhere, you really need to understand which rules are being applied to which cells. The rules tracer can help with this. Find one cell that you think holds the *wrong* number and look at how it is being calculated. I would guess it's not coming from where you think it is.

Going back to Martin's point, it seems likely that 'MR_2+10' and 'Actuals' aren't intended to be the same. I would re-visit this assumption.