Man, it is really worrying to me that a 20 minute video taught me a lot more than 3 hours in class with our professor...I feel like so much money is spend for nothing in the academic world
school is to weed out people who can follow instructions and good memory. TH-cam can sometimes teach you A LOT more than school. (I have a PhD and two masters and undergrad from top 10 university)
Such an art of teaching! Grateful to your video that saves me in both Econometrics and Data Science graduate classes. The series should be introduced to much more learners! Thank you so much!
The explanation on each detail really make me understood this topic even better. Every examples and detail really give the clear picture of this topic.
Nice and well explained video. Saved me a lot of time going through old text books where intuition about these concepts is often much more difficult to get. Keep up the good work dataminingcae!
I glad that I watch this video. The explanation really great, compact yet concise. As a university student its easy for me to understand the content. Thank you so much for your great work.
The explaination about the dummy variable as you have explained using more than one examples is really helpful. I manage to undertand the dummy variables rather than before.
The examples used in this video are easy to understand, plus with your interesting way of explanation. I am glad that this video exist. Thank you for this great video.
As a student this video is truely make me more understand. The explanation is very easy to understand with the example. Now I know on how to interpret more than two categories through this video.
From the video, now I know much better and understand how to interpret coefficients in regression analysis. The explanation is very well to understand by using graphics. Thank you sharing the input.
the example you give makes me more understanding about the dummy, especially on how to use it and interpret it. During this pandemic, your video is very suitable for us students to learn about the dummy variables.
By watching this video, it help me to understand more about what dummy variable is. It helps me to understand how to interpet the coefficient. Thank you for you explanation
The concept of dummy variables was very clearly explained. As a student, it's easy for me to understand the content after I watch this video. Thank you so much for helping me.
after watching this video, now i become more clearly about dummy variables and the way to construct and interpret their coefficient. thanks for this video and explanation. its also interesting.
Its look very hard to understand at first, but then when you come out with good example I manage to be more understand on how the dependent dummy variable changing depending categories. Thank you for very simple explanations .
It hard to understand on how to interpret when we have more than two categories, but then I manage to understand clearly when watch this video. I am glad to watch this video. It is very interesting video as it simple explanation that make me more understand.
It all about the dummy variable and i slowly learn what is dummy variable. How to use and and interpret it in the coefficient in a multiple or multivariate linear regression. It take time for me to understand it. However, i like your video because giving a simple example to make me understand more about it. Thanks in advance. :)
At first, it is complicated for me to understand on how to interpret when we have more than two categories, but then I manage to understand clearly based on your explaination. It is very interesting video with a great graphical example.
From this video, i can know how to interpret the coefficients in a multiple and multivariate regression because the lecturer using the simplest example to understand what actually he talking about. Thanks for this interesting video !
I found an easy way to interpret the coefficient when we have more than two categories. This video helps me a lot on understanding about dummy variables.
This video is so great! Before, I only understood this topic on the surface, after watching this, I now know how to further interpret the dummy variable better. This video also helped as reference for my assignments!
It is very useful for me due to the explanation and the example are close to me as a student. It make me figure out what dummy do and how to explained it.
I have difficulty to understand on how to interpret when we have more than 2 categories before, but with your useful explanation, i can understand it clearly now. Thank you for your explanation
i struggled understanding what is this dumb*** dummy variables are the entire semester and got 0 for interpreting the coefficient wrongly in mid term exam, thank god i found this video!!! make sense now on the comparing expected salary of treatment group to the control variable!!! thank you, you are awesome!!!!
From the video now I know how to make a solution and interpret dummy variables coefficients even in multiple or in multivariate linear regression . It helps me a lot in understanding the lecture given because it gave us some simple examples for us to understand
Thankyou so much for your teaching sir. You just saved my life I was very frustrating because I get confused more and more I couldn't wrap my head around all those terms and complex interpretation in my study and textbooks and need to do assignment. Now I have understood better and clear. Thankyou Thankyou Thankyou.
Thank you for such a fantastic video! The concept of dummy variables was very clearly explained. This video is much better than all others out there I have seen posted on a similar topic. Thanks again!!
Well done 👏🏻 I will note that the final model has an adjusted R squared suggesting that 28% of variability is explained by the IVs, which is interesting in and of itself.
This comment is an attempt at the "homework" problem assigned at the end. If you see anything wrong with my solution, then please feel free to correct me! Hopefully, this serves as a good guide on how to approach this problem. Salary = β0 + β1•Professional + β2•Assistant + β3•Associate + β4•Full + ε First, let's start with just distinguishing each binary dummy variable: If β1= 0, then the person is a professional; if β1= 1, then the person is an academic If β2 =0, then the person is not an assistant; if β1= 1, then the person is an assistant If β3 =0, then the person is not an associate; if β1= 1, then the person is an associate If β4 =0, then the person is not a full professor; if β1= 1, then the person is a full professor Following this reasoning, if all the dummy variables were 0, then the person would be an academic who is not an assistant, associate, or a full professor. Remember the general rule that if there is k categories, use k-1 dummy variables. Since there are four dummy variables, that means we are analyzing 5 categories (at least in the way that I understand it). Based on categories from the dataset, the non-dummy variable is instructor. Thus, if we set all of our dummy variables equal to zero, the coefficient of β0 would yield the salary of an academic instructor. Therefore, we can deduce the following equations, categories, and therefore salaries for five different ranks of professors based on the coefficient of β0 (an academic instructor). β0 = Academic Instrcutor = 81.2179 = $81,217.9 β0 + β1 = Professional Instructor = 81.2179 + (-22.7707) = 58.4472 = $58,447.2 β0 + β2 = Academic Assistant = 81.2179 + 27.0501 = 109.2680 = $108,268 β0 + β3 = Academic Associate = 81.2179 + 30.5468 = 111.7647 = $111,764.7 β0 + β4 = Academic Full (Professor) = 81.2179 + 57.0336 = 138.2515 = $138,251.5 Figuring out the salary of the last set of professors is a little tricky, but we can figure it out by just thinking about β1. Remembering that if β1 = 1, a person is a professional. A person being an assistant implies that they are not an associate or full professor. In general, you are either an instructor, assistant, associate, OR full (professor). No one can be two of these ranks at the same time; however, a person being a professional does not mean that they can't be an assistant, associate, or a full professor. Thus, we can deduce the following equations and therefore salaries for the last three ranks of professors (again, based on the salary of an academic instructor). β0 + β1 + β2 = Professional Assistant = 81.2179 + (-22.7707) + 27.0501 = 85.4973 = $85,497.3 β0 + β1 + β3 = Professional Associate = 81.2179 + (-22.7707) + 30.5468 = 88.994 = $88,994 β0 + β1 + β4 = Professional Full (Professor) = 81.2179 + (-22.7707) + 57.0336 = 115.4808 = $115,480.8 It's late so some of my logic is not explained very well, so if you have any questions just go ahead and ask! But assuming I didn't have some cataclysmic logical failure or arithmetic miscalculation, those are the average salaries of the 8 different types of professor positions. I am not certain that any of us would be able to test whether the salary differences between different ranks would be statistically significant without having the data ourselves to switch around the dummy variables.
You have written wrong when distinguishing dummy variables. If β1= 0, then the person is a academic; if β1= 1, then the person is an professional should be like this. The rest is correct
This really help me in understand for my ECO 545 subjcet chapter. I hope this understanding will help me to answer my test tommorrow on 30 june 2020. Wish me luck guys.
As indicator variables, as they will generally have a value of 1 to indicate if an observation belongs to a category and a 0 to indicate if not. They called the dummy because they are not real. This video is about 26, 253 professors salary in US. DV is salary and IV are qualification (academic&professional) and rank (assistant, associate,full&intructor).
Very good lecture. Things are clear and very well explained. A great job done by you dear faculty. I will like to search more lectures of you. Can you help??
It is hard understanding. But the explained was good to be the references . For example The problem with the ctegorical variable is what number should give to the different categories . I can more understand with repeating the video . Thank you
I got lost at the end. 22:50 You said we always have to take k-1, but in your home work example you included all of them? Is there a 5th category here?
Thanks for very simple explanation but How can we handle when we have around 20 Nominal and 20 ordinal categorical data . I face such problem in during Ames Housing data....? and finally if we use your method how we can perform feature selection...? because for that I have to consider whole column in your case (like qualification ) if particular levels of column are less significant then on what criteria we should select or drop that column
![[LIVE] : ONE ลุมพินี 89 | คู่เอก "ยอดไอคิว vs คิริลล์"](http://i.ytimg.com/vi/Z3vYkHJ_5Ig/mqdefault.jpg)
Man, it is really worrying to me that a 20 minute video taught me a lot more than 3 hours in class with our professor...I feel like so much money is spend for nothing in the academic world
school is to weed out people who can follow instructions and good memory. TH-cam can sometimes teach you A LOT more than school. (I have a PhD and two masters and undergrad from top 10 university)
@@MrSpiritmonger My god, this video answered all my questions that I couldn't figure out by watching our professors video over and over again.
How are you sure it is your professor and not your will power, focus and attention that made the difference.
Such an art of teaching! Grateful to your video that saves me in both Econometrics and Data Science graduate classes. The series should be introduced to much more learners! Thank you so much!
The explanation on each detail really make me understood this topic even better. Every examples and detail really give the clear picture of this topic.
This series of videos seems to be the best of its kind. Thank you!
Can't thank you enough prof! Really simple yet elaborated explanation! :)
Nice and well explained video. Saved me a lot of time going through old text books where intuition about these concepts is often much more difficult to get. Keep up the good work dataminingcae!
Best video I have encountered on this topic. Thank you!
I glad that I watch this video. The explanation really great, compact yet concise. As a university student its easy for me to understand the content. Thank you so much for your great work.
Simple, crisp and made is very easy to understand. Thanks
The explaination about the dummy variable as you have explained using more than one examples is really helpful. I manage to undertand the dummy variables rather than before.
This is one of the best explanation for dummy variable. Thank you so much for your help and kindness. You are the best professor
This is the greatest explaination. Thank you so much for your help and kindness. You are the best professor
The explanation with example really help me to understand more on this chapters. Thank you
The examples used in this video are easy to understand, plus with your interesting way of explanation. I am glad that this video exist. Thank you for this great video.
The best video I have ever watched about the dummy variables! Thanks a world!
As a student this video is truely make me more understand. The explanation is very easy to understand with the example. Now I know on how to interpret more than two categories through this video.
From the video, now I know much better and understand how to interpret coefficients in regression analysis. The explanation is very well to understand by using graphics. Thank you sharing the input.
From the video i understand how to interpret the two categories. Also the problems of variables. Thank you so much for this knowledge. 👍
The explanation is precise and great. It make me more understand about dummy variables. Thank you so much for this great work.
the example you give makes me more understanding about the dummy, especially on how to use it and interpret it. During this pandemic, your video is very suitable for us students to learn about the dummy variables.
By watching this video, it help me to understand more about what dummy variable is. It helps me to understand how to interpet the coefficient. Thank you for you explanation
The concept of dummy variables was very clearly explained. As a student, it's easy for me to understand the content after I watch this video. Thank you so much for helping me.
Explanation using some example are exactly the way to make me more understand 😍Now I'm getting better in understanding dummy variables. Good job 👍🏻
after watching this video, now i become more clearly about dummy variables and the way to construct and interpret their coefficient. thanks for this video and explanation. its also interesting.
Its look very hard to understand at first, but then when you come out with good example I manage to be more understand on how the dependent dummy variable changing depending categories. Thank you for very simple explanations .
It hard to understand on how to interpret when we have more than two categories, but then I manage to understand clearly when watch this video. I am glad to watch this video. It is very interesting video as it simple explanation that make me more understand.
It all about the dummy variable and i slowly learn what is dummy variable. How to use and and interpret it in the coefficient in a multiple or multivariate linear regression. It take time for me to understand it. However, i like your video because giving a simple example to make me understand more about it. Thanks in advance. :)
At first, it is complicated for me to understand on how to interpret when we have more than two categories, but then I manage to understand clearly based on your explaination. It is very interesting video with a great graphical example.
An excellent explanation of this subject covering all aspects! Thanks.
For this video, it shows and helps how to interpret coefficient with two categories
From this video, i can know how to interpret the coefficients in a multiple and multivariate regression because the lecturer using the simplest example to understand what actually he talking about. Thanks for this interesting video !
I think i more understand about this topic thanks to your good explanations. Its very interesting and not boring at all.
Your explanation so clear..it easy to understand. Thank you for sharing. It help me a lot..
Best video ever seen on this topic..
I found an easy way to interpret the coefficient when we have more than two categories. This video helps me a lot on understanding about dummy variables.
This video is so great! Before, I only understood this topic on the surface, after watching this, I now know how to further interpret the dummy variable better. This video also helped as reference for my assignments!
Thank you so much for this! You made the concepts easy to understand.
Your explanation was very helpful. Before this I have problem on how to interpret the coefficients but now I'm bit understand compare than before.
It is very useful for me due to the explanation and the example are close to me as a student. It make me figure out what dummy do and how to explained it.
I have difficulty to understand on how to interpret when we have more than 2 categories before, but with your useful explanation, i can understand it clearly now. Thank you for your explanation
Thanks a lot for this video and for the other ones, it helped me a lot!
good video because he keeps me more thinking. good information.
i struggled understanding what is this dumb*** dummy variables are the entire semester and got 0 for interpreting the coefficient wrongly in mid term exam, thank god i found this video!!! make sense now on the comparing expected salary of treatment group to the control variable!!! thank you, you are awesome!!!!
Simple and concise, easy to understand! Thank you!
From the video now I know how to make a solution and interpret dummy variables coefficients even in multiple or in multivariate linear regression . It helps me a lot in understanding the lecture given because it gave us some simple examples for us to understand
Thank you for making it so simple to understand.
Thank you for such clear explanaitions!
Thankyou so much for your teaching sir. You just saved my life I was very frustrating because I get confused more and more I couldn't wrap my head around all those terms and complex interpretation in my study and textbooks and need to do assignment. Now I have understood better and clear. Thankyou Thankyou Thankyou.
This video really help me to understand about dummy variabe and how to interpret these coefficient. :)
Fantastic lesson. Thank you so much. It helped me immensively
thanks ! great explanation about dummy variables and it helps me a lot as a student
Amazing, awesome, fantastic.... are very small words for your videos..... I just love you!
Awesome example and explaination👍
This video really make me understand how to interpret the coefficient. Really usefull. Thank you
Thanks - this was a really clear explanation of dummy variables.
Kudos to you dear we appreciate your efforts. thanks 👍👍👍 👍
Thank you for such a fantastic video! The concept of dummy variables was very clearly explained. This video is much better than all others out there I have seen posted on a similar topic. Thanks again!!
Thank you for your explanation. I hope it help me to answer the test question. This video help me a lot how to understanding dummy variable
my brain says cool. soft voice & good way of explination, all the best for your career, try to explore more
Well done 👏🏻 I will note that the final model has an adjusted R squared suggesting that 28% of variability is explained by the IVs, which is interesting in and of itself.
Thank you so much, this is what im looking for. I can refer to this video for additional info !
your teaching is so clear, thanks a lot!
thanks to You sir. i feel humbled you are even thanking us.
Thank you so much! I really needed this for my thesis!
this is an excellent video !! thank you prof.
thank you for teaching everyone.
Excellent explanation. Thank you so much
This comment is an attempt at the "homework" problem assigned at the end. If you see anything wrong with my solution, then please feel free to correct me! Hopefully, this serves as a good guide on how to approach this problem.
Salary = β0 + β1•Professional + β2•Assistant + β3•Associate + β4•Full + ε
First, let's start with just distinguishing each binary dummy variable:
If β1= 0, then the person is a professional; if β1= 1, then the person is an academic
If β2 =0, then the person is not an assistant; if β1= 1, then the person is an assistant
If β3 =0, then the person is not an associate; if β1= 1, then the person is an associate
If β4 =0, then the person is not a full professor; if β1= 1, then the person is a full professor
Following this reasoning, if all the dummy variables were 0, then the person would be an academic who is not an assistant, associate, or a full professor. Remember the general rule that if there is k categories, use k-1 dummy variables. Since there are four dummy variables, that means we are analyzing 5 categories (at least in the way that I understand it). Based on categories from the dataset, the non-dummy variable is instructor.
Thus, if we set all of our dummy variables equal to zero, the coefficient of β0 would yield the salary of an academic instructor. Therefore, we can deduce the following equations, categories, and therefore salaries for five different ranks of professors based on the coefficient of β0 (an academic instructor).
β0 = Academic Instrcutor = 81.2179 = $81,217.9
β0 + β1 = Professional Instructor = 81.2179 + (-22.7707) = 58.4472 = $58,447.2
β0 + β2 = Academic Assistant = 81.2179 + 27.0501 = 109.2680 = $108,268
β0 + β3 = Academic Associate = 81.2179 + 30.5468 = 111.7647 = $111,764.7
β0 + β4 = Academic Full (Professor) = 81.2179 + 57.0336 = 138.2515 = $138,251.5
Figuring out the salary of the last set of professors is a little tricky, but we can figure it out by just thinking about β1. Remembering that if β1 = 1, a person is a professional. A person being an assistant implies that they are not an associate or full professor. In general, you are either an instructor, assistant, associate, OR full (professor). No one can be two of these ranks at the same time; however, a person being a professional does not mean that they can't be an assistant, associate, or a full professor. Thus, we can deduce the following equations and therefore salaries for the last three ranks of professors (again, based on the salary of an academic instructor).
β0 + β1 + β2 = Professional Assistant = 81.2179 + (-22.7707) + 27.0501 = 85.4973 = $85,497.3
β0 + β1 + β3 = Professional Associate = 81.2179 + (-22.7707) + 30.5468 = 88.994 = $88,994
β0 + β1 + β4 = Professional Full (Professor) = 81.2179 + (-22.7707) + 57.0336 = 115.4808 = $115,480.8
It's late so some of my logic is not explained very well, so if you have any questions just go ahead and ask! But assuming I didn't have some cataclysmic logical failure or arithmetic miscalculation, those are the average salaries of the 8 different types of professor positions.
I am not certain that any of us would be able to test whether the salary differences between different ranks would be statistically significant without having the data ourselves to switch around the dummy variables.
You have written wrong when distinguishing dummy variables.
If β1= 0, then the person is a academic; if β1= 1, then the person is an professional
should be like this. The rest is correct
This really help me in understand for my ECO 545 subjcet chapter. I hope this understanding will help me to answer my test tommorrow on 30 june 2020. Wish me luck guys.
Thank you for your explanation 👍, now i can understand better
Very nice explanation. My brain was hanging up on the meaning of the coefficients, but the idea seems obvious now.
Wauw! This video is absolutely great. Good job!
Thank you for the video. I think, I am more understand with dummy variables and why we used of 0 and 1 at the e-views data.
Thanks a ton! This video was extremely helpful
As indicator variables, as they will generally have a value of 1 to indicate if an observation belongs to a category and a 0 to indicate if not. They called the dummy because they are not real. This video is about 26, 253 professors salary in US. DV is salary and IV are qualification (academic&professional) and rank (assistant, associate,full&intructor).
9 years later your video is still giving tremendous help to students, thank you for your work.
Very good lecture. Things are clear and very well explained. A great job done by you dear faculty. I will like to search more lectures of you. Can you help??
Exceptional work !!
Hey man. Just wanted to let you know my Uni links to your vids as preparation, and you do them very well. Thanks for everything!
It is hard understanding. But the explained was good to be the references . For example The problem with the ctegorical variable is what number should give to the different categories . I can more understand with repeating the video . Thank you
Very clear and useful... Thanks a lot
so helpful omg too certi broski love for that my g
Can't thank enough. I do understand more on how to intrepret the coefficient when there are more than two catagories
This video basically similar with previous video but further discussion on differences between continuous and categorical variables in regression.
dude you are amazing. 10 out of 10
thanks you have cleared all ambiguities
Great explanation but you didn’t explain the dummy variable trap. That was the part I was especially waiting for .
hi why do coefficients of other variables change when you include dummy variables.
I got lost at the end. 22:50
You said we always have to take k-1, but in your home work example you included all of them? Is there a 5th category here?
tks so much! That's what i'm looking for.
thank you for the amazing video
Great one. Is it possible tovpost the data set as well?
Here's the link: figshare.com/articles/dataset/Salaries_Professor_in_American_Universities/15138300
Thanks for very simple explanation
but How can we handle when we have around 20 Nominal and 20 ordinal categorical data . I face such problem in during Ames Housing data....? and finally if we use your method how we can perform feature selection...? because for that I have to consider whole column in your case (like qualification ) if particular levels of column are less significant then on what criteria we should select or drop that column
Where can we find the answer for the questions asked in the end?
Can you please help me with where I would find a dataset?
Why can't we have one regression equation for each different categorical variable and then substitute them
Great video, so clear
It give me input about regression. Thank you
the last bit was what i needed
wow..thanks you so much for this video