Deneb Normal Distribution using Power BI dataset mean and standard deviation

[Greg]

Hi @nithesh.n15

The original thread has not yet been split and a new thread has not yet been created, so I’m creating a new thread and linking to the original thread.

Deneb Example - Distributions

Deneb/Vega-Lite has built-in “binning” ability and can create histograms from raw data to display distributions. Further, while Power BI only permits a single dataset to be provided to custom visuals, additional datasets can be created using Vega-Lite’s data generators. The example herein consists of a 20-bin histogram of IMDB movie ratings overlaying a parameterized normal distribution using data from a sequence generator. This example illustrates a number of Deneb/Vega-Lite feature…

Q (as it was added to original solved thread):
Is it possible to change the distribution chart like Normal distribution’s mean to be as IMDB Rating’s mean as default and not to allow user’s to modify the mean. Currently here user can change the mean of Normal distribution.

A:
Yes.

The statistics were added to the Power BI dataset instead of calculating them directly in Deneb/Vega-Lite. Two new measures were added:

Mean = CALCULATE( 
    AVERAGE( Movies[IMDB Rating] ), 
    REMOVEFILTERS( Movies) 
    )

Standard Deviation = CALCULATE( 
    STDEV.P( Movies[IMDB Rating] ), 
    REMOVEFILTERS( Movies ) 
    )

Then in Deneb/Vega-Lite, 2 “lookup” transforms were added to add the [Mean] and [Standard Deviation] fields to the [Sequence] dataset. Finally, the [densityNormal] calculation was revised to use the [Mean] and [Standard Deviation] from the dataset.

(The proper calculation of the value [_y_factor] for the normal distribution is left for further development; I used a value of 1000 in this example.)

Hope it helps.
Greg
eDNA Forum - Deneb Distributions - V1.pbix (1.5 MB)

[nithesh.n15]

Thank you Sir. You are awesome.