6 Replies Latest reply: Oct 24, 2013 8:25 AM by Arnaud Klr

# Chart with a moving average

Hey everyone,

I woule like to create an chart that is showing the moving average of completness (in %) of my task, and i don't find the expression needed to use a moving average.

For exemple of moving average in my source file, i have:

1/1/20134580%80%
1/2/20133475%77,5%
1/3/201361060%71,67%
1/4/2013121580%73,75%
1/5/201322100%79%

In this way, i would like to have a line chart showing the moving average evolution by date.

I don't know if it's clear, but it will be really nice if someone can help me on this tricky chart

Best regards,

Arnaud

• ###### Re: Chart with a moving average

Hi,

See the attached app,

• ###### Re: Chart with a moving average

Dear Marie-Sophie,

Thank you for your helpfull example, i think i understand the formula.

How would you implement this in a chart where the Dimension chosen is the date, and my current expression is:

"Count(Distinct if(Action_Date >= Action_Date_Close,ID_Planning)) / Count(DISTINCT ID_Planning)"

ID_planning is an unique code for every task.

Thank you so much in advance, i'm pretty new on Qlikview, and it's a bit tricky at the start

Best regards,

Arnaud

• ###### Re: Chart with a moving average

Could you post your qvw (or part of it, or even an application with false date ..) ?

• ###### Re: Re: Chart with a moving average

Dear Marie-Sophie,

Here is the Qlikview with an example of the graphic where i want to change the expression, in a moving average expression.

Best regards,

Arnaud

• ###### Re: Re: Re: Chart with a moving average

Hi,

I cannot upload files right now but you can maybe create a second expression with this formula :

Rangesum(Top([% of Recurrent tasks Completed], 1, RowNo()))/ rowno()

Then you unchek line and symbol for your first expression and check it for the new one.

The only problem with this formula is that lines removed by your if statement are counted in the moving average.

I'll try to find a better solution.

• ###### Re: Re: Re: Chart with a moving average

Thank you a lot for your time !