Forum Discussion
Slope and graph plot
Hi Courju
sorry for not very understand,
based on the file I created, could you add more details about the expected result?
Best Regards,
Community Support Team _Tang
If this post helps, please consider Accept it as the solution to help the other members find it more quickly.
Thanks for already starting to look at my request ... 😁
I wanted to make changes to your pbix, but somehow couldn't.
But, let me show it then, as you proposed:
Based on what you started:
I have a huge database from production samples. After slicer filtering down to one single property, I end up with a similar table than the one you proposed.
No average necessary, no sum, and but a numerical 1.2.3.4. is given automatically (of course, as it needs to follow the slicer choice, so it can't be just added to the database). If I do add an ID to each row, it would not be continuous, due to the lack of data on certain days, and for certain properties.
The plot is done of the numerical vs. day (no double point will be found as my values are refered with day and time and for one property, there is no double value then) and the linear regression done provides me the 0,3281. This is divided by the average of all data to rationalize the result. There I get 3,69% "increase".
Purpose is to obtain a simple visualisation like this:
Based on, let say 5%.
For a quality control lab, it will be a visualization for the manager to look at every 2-3 days. No graph, no values, only lights red or green, showing that the property is stable or not, but always based on the same rational increase.
Of course, a property measuring at average 0.62 is not impacted the same from a +1 unit than a property measuring at average 10500 ... that is clear.
Ok, I hope this is easier to understand now.
Thanks again for your help ... already hours trying that ...
Julien