The New LINEST() DAX Function in Power BI - Regression Analysis Made Easy

Рет қаралды 7,669

Жыл бұрын

The new DAX function LINEST() is a new powerful tool in your Data Analysis skillset.
It is the easiest way to do Regression Analysis. And Regression Analysis is a fundamental concept in Data Science.
LINEST() function uses the Least Squares method to calculate a straight line that best fits the given data, then returns a table describing the line.
The equation for the line is of the form: y = Slope1*x1 + Slope2*x2 + ... + Intercept.
While we had this function in Excel for many years, only with this month update we have it in Power BI.
====CHAPTERS====
01:23 LINEST() documentation
02:40 LINEST table creation
03:20 Equation for linear regression
04:30 Parameters table (years)
07:07 Measure for selected years
08:13 2nd Equation for prediction
Here is the link to the pbix file:
drive.google.com/file/d/1VLkY...
The dataset used in this tutorial is called Salary_Data from Kaggle.com
#dax #powerbi #linest #dataanalysis #businessintelligence #linearregression #regressionanalysis

Пікірлер: 19

@angelmaravilla4708 Жыл бұрын

Man, I just want you to know, that I've just found your channel, I've just watch your video, hit the dumb buttom, suscribe to EVERY notification, and share your content with my people. Really thanks for this amazing content; keep breaking it 🤙

@Data.Analytics.Central Жыл бұрын

Dear Angel, messages like yours motivate me to keep publishing this kind of practical videos. Many thanks for the appreciation! 👍

@yes4meliorism830 Жыл бұрын

Very well explaned.

@viorel1316 Жыл бұрын

Thank you for the appreciation! 🙏

@user-yl1fo5wy6q 4 ай бұрын

Thank you

@Data.Analytics.Central 4 ай бұрын

I am glad that you have found it useful!

@Cirocarbone 10 ай бұрын

great! very helpful video, thank you! My "Salary_Data" table is similar but also has a third column (object) which I need to filter in. The code to calculate SalaryEST, in fact, computes all mixed ebjects, even if I select a single object by a slicer from Salary_Data. Is there a simple solution without making X,Y table for each single object in Salary_Data?

@rafaelfelipenovi6308 Жыл бұрын

Thanks for amazing video, If i have a column 'product' how i do a equation for each product? Because only did to all together.

@rafaelfelipenovi6308 Жыл бұрын

I did this with r script in power bi but i trying in dax too

@Data.Analytics.Central Жыл бұрын

Hi Rafael. Thanks for the feedback 🙏 Very good question, this will be the topic of the next video. Stay tuned!

@Data.Analytics.Central Жыл бұрын

kzfaq.info/get/bejne/htFih5BznK3Repc.html

@p.j.882 Жыл бұрын

Missing items: standard errors values of coefficients, coefficient of determination, standard error of y estimate, f statistic, degree of freedom, regression sum of squares, residual sum of squares.... check excel help on linest.

@Data.Analytics.Central Жыл бұрын

Hello, those parameters are to be found in the LinestResult table. The scope of the video is only to show how the Linest function works in practice, as its documentation is quite scarce (eg. they do not mention to wrap the Slope and Intercept inside a SUM or a MAX in the regression equation.

@pjarina Жыл бұрын

@@Data.Analytics.Central ...no offence., but LINEST in excel, google sheets, open office, and libreoffice has those statistics

@Data.Analytics.Central Жыл бұрын

@@pjarina have you watched the video? At 3:03 you have the list of the statistics that you have mentioned. It gives the same statistics as in Excel

@pjarina Жыл бұрын

@@Data.Analytics.Central ...yes but the missing items are not defined.