How to add new plots based on check box filters

[Anonymous]

My data has 4 columns: start_date, variant, level, percentage.

The x-axis is the level.

The y-axis is the percentage.

The legend is the variant.

The check box filter is the start_date.

The plot is a line chart sort by the level in ascending order.

For each start_date, I have variants, levels, and percentages.

Here's what I want: For each ticked start_date, I want to add its corresponding set of variant-level-percentage to the existing line chart that I've created, with a new set of legends being created. For example, if I tick 2021-Jun-01, it should show all the variants-levels-percentages on the line chart. If I add 2021-Jul-01, it should add a new set of legends to the existing legends and basically show another set of variants-levels-percentages on the same line chart. So if the 2021-Jun-01 has 4 variants and 2021-Jul-01 has 3 variants, then I would have 7 variants in my legend, which means I would have 7 lines on my line chart.

How do I achieve this?

[TomMartens · Accepted solution]

Hey @Anonymous ,

the description you provided should exactly return what you would expect.

The legend items are represented by lines, if there are 7 distinct legend items, there will be 7 lines.

Of course, if there are only 6 distinct legend items across 2 start dates, there will be 6 lines, as one item will be represented by an aggregation of numeric values.

This is how it works and can not be resolved.

Maybe you can use small multiples to keep the legend items separated, the next screenshot demonstrates this:

The chart is using ColorName on the Small multiples bucket and at the same time, the column ColorName is used as a source for the slicer. In this example, the column ColorName is playing the part of your start_date column

Be aware that the Small multiples feature is in preview, this means you maybe have to enable the feature in the options settings. This article provides more information on how to use this feature: Create small multiples in Power BI (preview) - Power BI | Microsoft Docs

Hopefully, this provides what you are looking for.

Regards,

Tom 

[amitchandak]

@Anonymous , You need to use an independent date table for that in slicer

then try a measure like this , assuming percentage is measure

Measure =

var _tab = summarize(filter(Table, Start_date in allselected(Date[Date])),variant)
return
calculate([percentage], filter(Table, Table[variant] in _tab))

refer why you need this

Need of an Independent Table in Power BI: https://youtu.be/lOEW-YUrAbE

Need of an Independent Date Table:https://www.youtube.com/watch?v=44fGGmg9fHI

[Anonymous]

No, percentage is not a measure. It's just a numeric data column that comes with the data import.