Set Analysis problem with ignoring selections

[#1]

I made a demonstration for Set Analysis problem. Demonstration file was added as attachment. .

when i click on specific year on the year table, I'm getting the trend graph for this year but i just want to see general graph from 2010-2012. I try "=count({1<Year=,V_ID={'=Actual>=Target'}>} V_ID)/Count({<Year = >} V_ID)" this formula. "V_ID={'=Actual>=Target'}"  works very well but "{1<Year=>}" doesn't work. In my opinion, "{1<Year=>}" must ignore Year selection but it doesn't work.

Thank you for your help

[Luis_Cortizo]

Hi, Selim

   For what I can see, the expression is working I supposed:

- On the first expression, whenever you click on a Year the values for V_ID change and the output of the expression is correct.

- On the mini chart, the expressions is working BUT the dimension of the MiniChart is Year as well, so whenever you click on a specific value of Year the dimension of the mini-chart gest filtered and if the mini chart only has one value the output will be one value only independently of the output of the expression.

Hope it helps.

Best regards.

[MK_QSL]

Can you explain how you want to compare V_ID with Actual and Target?

V_ID={'=Actual>=Target'}

[#4]

Thank you for your explanation.

Is there a way to show on the first expression, "cal",  based on a selected year, on the second  expressions ,"trend", as a mini chart not year dependant ?

[#5]

Hi Manish,

it's not a comparison.It was used for obtaining to return records having target data greater than or equal to actual data. It woks

[Luis_Cortizo]

I can't think of any way of doing it right now (beyond modifying the data model).

The mini chart needs a field for the dimension value and it cannot be altered or calculated, so whenever you select a year the dimension of the mini-chart will be filtered.

[preminqlik]

hi there once try this

count({<Year=>}aggr(count({1<V_ID={'=Actual>=Target'}>}V_ID)/Count( {<Year=>} V_ID),D_ID,Name,Year))

[#8]

Dimension ignoring is working but i think formula is wrong.