Dr. Himayatullah, Its really awesome the way you depicted in a very succinct manner with a very simple example and anyone can easily understood for the whole life. I am PhD Scholar (Machine Learning, Artificial Intelligence) and really stuck in multicollinearity problem. Thanks & regards,
Great shiv. You seem to have understood. Now, what will you do with the collinear variables? Everybody explains till VIF but don't explain what to do with the collinear variables.
@@HKofficial15 Thank you so much for your reply. I am aware of handling collinearity in data sets. I was mainly mentioning it in favor of other viewers of your videos. It might benefit them, if you could please post a video on how to handle them. Since you have such prolific experience in teaching and nice youtube tutorials, other students may benefit from learning the 3 different ways of handling multicollinearity in datasets. I do not have a channel where I could teach but you have a fantastic platform where you can address this critical issue that researchers are faced with often. I would appreciate your help. Thanks. Great Job.
Your aim shouldn't be to improve these values. You major focus must be on the algebraic signs of the coefficients of expl variables and their significance. Values of R and Rsquared matter not the important. Your interpretation of the results must be relevant and correct.
i want to ask a question not about this video it's about descriptive statistics please i need to know the answer why is it important to know the mean, the median, minimum, maximum and the standard deviation? why do we have to know these things? what difference does it make if the mean of x1 for example is 1.7 or it is 25.9 or for x2 the mini. is 0.000 and the max is 117.5. i really need to know the answer. thank you
dear Crochet mean, median, mode are a measure of the centeral tendency of a given data when we say this it means how your data are close to the center. tells you whether your data are normally distributed or not whereas standard deviation, range, a coefficient of variation measures how your data is far from the center or how your data is dispersed. mean =1.7 and 25.9 it is not possible to tell the difference without looking at your data distribution but in the case of min and max from 0.000 to 111.75 it tells your data is highly dispersed in which we will have a highly standard deviation
Unfortunately, you couldn't logically and technically remove the multicollinearty problem. You just increased the sample size. What if we have a fixed sample size? For example, 50 states in U.S. Can you increase the sample size now ? U.S is 50 states and if you add even one or two more states that would mean you are trying to change the U.S. map on the panorama ! Does it make sense?
Dear... presence of multi-collinearity will always be there. But it is the degree or severity of collinearity that matters. USA-50 states is an exceptional case. But in other cases you can increase the sample size. That may lightens the severity of the multicollinearity.
My sample size is fixed so I cnat change the size and reduce the multicollinearity so can you please guide to get rid of multicollinearity in another ways..
Dr. Himayatullah, Its really awesome the way you depicted in a very succinct manner with a very simple example and anyone can easily understood for the whole life. I am PhD Scholar (Machine Learning, Artificial Intelligence) and really stuck in multicollinearity problem.
Thanks & regards,
Thanks Raza Abidi Sahib. Its my try to make it as simple as possible for those who have problems with issues like this.
Wonderful sir....So clearly understood the concept. Thank you..
Welcome Sivakumar
Great shiv. You seem to have understood. Now, what will you do with the collinear variables? Everybody explains till VIF but don't explain what to do with the collinear variables.
clearly understood refer this video thank you sooo much sir
Thanks Nirodha...Its a pleasure for me if you and other students could benefit from this,
Best video on RUclips on this topic. Thank you very much!
thanks dear..I always try to make simplest possible presentations
Thanks for your encouragement
Can you now explain how to handle those collinear variables. That's where most students get stuck. Thanks. Nice tutorial.
Try someway to resolve the issue. Check the VIF and tolerance of all multicollinear variables. The variable having the highest VIF must be removed.
@@HKofficial15 Thank you so much for your reply. I am aware of handling collinearity in data sets. I was mainly mentioning it in favor of other viewers of your videos. It might benefit them, if you could please post a video on how to handle them. Since you have such prolific experience in teaching and nice youtube tutorials, other students may benefit from learning the 3 different ways of handling multicollinearity in datasets. I do not have a channel where I could teach but you have a fantastic platform where you can address this critical issue that researchers are faced with often. I would appreciate your help. Thanks. Great Job.
@@gannechaitanya6703 I appreciate your query it is important for students to think on that line. Thanks again.
Perfectly explained
I am happy that it helped you
Welcome Uzma..I have some new videos uploaded..you may check out
Sir i am facing the problem with r value and r2 value kindly guide me as how to improve the mentioned value.
Your aim shouldn't be to improve these values. You major focus must be on the algebraic signs of the coefficients of expl variables and their significance. Values of R and Rsquared matter not the important. Your interpretation of the results must be relevant and correct.
Sir can it be possible for you to share your mail id as I m getting lost of confusion regarding the data interpretation.
Regards
Anjali
@@HKofficial15 kindly explain it
@@HKofficial15 sir kindly explain the correct steps for hypothesis testing specifically when my aim is to use SEM.
i want to ask a question not about this video it's about descriptive statistics please i need to know the answer
why is it important to know the mean, the median, minimum, maximum and the standard deviation? why do we have to know these things? what difference does it make if the mean of x1 for example is 1.7 or it is 25.9 or for x2 the mini. is 0.000 and the max is 117.5. i really need to know the answer.
thank you
dear Crochet
mean, median, mode are a measure of the centeral tendency of a given data when we say this it means how your data are close to the center. tells you whether your data are normally distributed or not whereas standard deviation, range, a coefficient of variation measures how your data is far from the center or how your data is dispersed. mean =1.7 and 25.9 it is not possible to tell the difference without looking at your data distribution but in the case of min and max from 0.000 to 111.75 it tells your data is highly dispersed in which we will have a highly standard deviation
Unfortunately, you couldn't logically and technically remove the multicollinearty problem. You just increased the sample size. What if we have a fixed sample size? For example, 50 states in U.S. Can you increase the sample size now ? U.S is 50 states and if you add even one or two more states that would mean you are trying to change the U.S. map on the panorama ! Does it make sense?
Dear... presence of multi-collinearity will always be there. But it is the degree or severity of collinearity that matters. USA-50 states is an exceptional case. But in other cases you can increase the sample size. That may lightens the severity of the multicollinearity.
My sample size is fixed so I cnat change the size and reduce the multicollinearity so can you please guide to get rid of multicollinearity in another ways..
i have used 386 samples and there still exists problem sir
Check your data thoroughly...there must be some issue in your data.
@@HKofficial15 wat mistake night have happen ..if you could explan once
Thanks sir
welcome
Nice content, I also have rural development and planning volume 1-3 on my RUclips channel
2:38