3. Interpreting Linear Regression Outputs in SAS

[@009kishor]

Great fan of your teaching style , learnt most of stats concepts from ur videos..

[@lw6858]

40:00 R SQUARED CAN BE INTERPRETED AS THE CORRELATION SQUARED AND ALSO AS THE PERCENTAGE OF THE SUM OF ALL VARIANCES THAT IS EXPLAINED BY THE MODEL 17:45 . YOU ARE A GOOD TEACHER SIR.

[@lw6858]

9:00 the coefficient of variation is the Root mean square error as a percentage of the dependent mean.

[@NiteshKumar-ny4zk]

Very helpful videos. Please release something on GLM too.

[@carlosdevia3459]

super helpful!

[@lovebooksandtea]

this is so helpful, thank you!!

[@rrk10786]

Hi, a small correction needed at 27:43 mins in the video.... R square formula explained should be (1-(10517/(6948+10517)). Great video though to understand the key concepts.👌

[@snprasad7605]

Excellent video.

[@ravindarmadishetty736]

R Square is not actually derived from "r" value. Generally if we square the values of r, we get r-square. But it is actually derived from Sum of Squares of Regression(SSR)/Sum of squres of Total(SST) from Anova table. Anova table is very important for linear models not only for r-square

[@appagi]

Hello Sir, you are simply awesome. Excellent explanation about every thing. To the point and just what is needed. _/\_ _/\_

[@inner5598]

very helpful. Thanks. Can you share common link for other videos ?

[@LearnanalyticsIn]

You can refer to the channel's playlist page for Linear Regression series and more.

[@vinaykarna5825]

You explained Coeff Var as ratio of Root MSE / Dependent mean, which is not the case here. 52.645/7.28 happens to be 7.23 approx. Now please can you explain what exactly is Coeff Var and how does it help in estimating linear regression? Thanks in advance.

[@LearnanalyticsIn]

The other way around, Root MSE is 7.28 and Dependent mean is 52.6, so Coefficient of Variation has to be calculated as 7.28/52.64 and not the inverse as you have done above.

[@vinaykarna5825]

Got it, sir. Crystal clear now.