Econometrics // Lecture 2: "Simple Linear Regression" (SLR)
An Introduction to the "Simple Linear Regression" (SLR) in Econometrics. This video covers:
1. A formal introduction to the SLR model
2. The difference between population and estimation models
3. A basic interpretation of the slope and intercept
4. What causality means
5. A more formal visual representation of the simple linear regression
6. Introduction to residuals
7. An outline of how to estimate the slope and intercept and where it originates from
Note: All of this applies to the "Ordinary Least Squares" (OLS) Estimation.
This video is to serve as a basic introduction to the "Simple Linear Regression" model. The video briefly touches on lots of subjects to ensure that the student gains a strong foundation for more in depth analysis to come.
Additional Comments:
If you want to estimate any ui, find the estimates for the intercept and slope and plug them into the ui equation: ui = yi - yi_hat = yi - (beta0_hat) - (beta1_hat)(xi). Additionally, remember that the derivative of y in respect to x represents the change in y as a result of a change in x. Therefore if we have a causal relationship, if x increases by 1, y will increase by Beta_1. This will be shown in depth in a later video.
The next video tutorial on "Ordinary Least Squares" and "Goodness Of Fit": • Econometrics // Lectur...
All video, commentary and music is owned by Keynes Academy.
Пікірлер: 138
This was actually very helpful. Definitely better than cramming 20pages from the book to explain one little thing like 'causal effect' Thank youu! :)
12:25 gee mate, I'm crying too... Thanks for the lecture, and for your sacrifice- was very insightful!
@evansiddique3775
4 жыл бұрын
Lmfao
@joebrodenberg3428
3 жыл бұрын
he didn't cry, he hiccupped
it is a great class, BUT I wish you can include some examples to clear the lecture. Thank you
Thank you for the video. Really wish you continued on with the series. Please upload more in the future.
Remember that the derivative of y in respect to x represents the change in y as a result of a change in x. Therefore if we have a causal relationship, if x increases by 1, y will increase by Beta_1. This will be shown in depth in a later video.
@grinch061
3 жыл бұрын
Why did you stop? Loved it
Thanks for the upload, you did a good job of explaining something that's quite complicated. Cheers.
Thanks man! Got my first in course exam coming up and this helped so much!
you explain everything so well. I wish you really go in dept in Econometrics!
It is an amazing explanation , please keep it on doing such very important works
Hi Ali, Thank you for the feedback! I will be posting more videos very shortly. If you need help with a specific topic, feel free to comment or message me, and I will post a video as soon as possible. If you are planning on enrolling in an introductory econometrics course at the undergraduate level, I would recommend "Introductory Econometrics: A Modern Approach" by Wooldridge. Good luck!
@michaelgondwe-oe5gd
2 ай бұрын
Continue making more videos , on probit and logit, log linear model, instrumental variables etc
Very nice presentation. I wish you made more videos on this series. Your voice is very soothing and clear to follow.
Thank you so much🌼 Please continue with this series sir✨ Really very helpful
This explanation was very easy to understand and helpful. Please upload more!
Good explanation thanks. Please continue to make these!
Thanks for the great videos. It helped me refresh my Econometrics. I think if you use some numbers to your examples it would me more intuitive to great number of people who are interested to the topic. Appreciate your time and effort.
thank you your explanation is much better than my lecturer. u made it easy to understand
Can you post more videos. Your Channel is saving my life!!! I love it!!
@studyzone131
4 жыл бұрын
Hahaha... how it possible ??
Great..Helped me a lot!!!! Keep Posting More Videos
You are amazing man! keep it up
A shame that you make only three of them, they was the most useful video for my econometrics exam
You are amazing...thumb's up for you.
Awesome video !
Thank you, Sir, Your explanation is very easy to understand. You made complicated things become simple. What a great job!
This is so helpful! Thank you so much!
Please continue to post videos!!!
why did you stop making videos???!!! You explain super good!
Amazing explanation. You should not have stopped making such videos.😞
very educational and useful video! You explained more than professor)
Hi Llamskid, we just posted our next video on OLS and Goodness-Of-Fit. Hope you enjoy! More to come!
Helped a lot. Thanks !
Thanks for the video!
great video simple to grasp
you have literally saved my ass! thanksssss so much! greets from Vienna :D
really good! thank youuuu
Hello, beforehand thanks for yor explanations but how can we find "beta one bar" could you show the calculations of this?-- thanks beforehand!
informative and helpful .please load some more videos.
Nice video, thanks!
Wow at 5:00 you explained something that I had copied down as notes but could not figure out what it meant! Goes to show that notes are only half the lesson
The equation at 9:36 should be : Ÿi = ûi + Yi
I definitely need to have a better understanding of the first video before watching this one comfortably
Big help man, thanks
You are a life saver♥️♥️♥️♥️♥️♥️♥️♥️♥️♥️♥️
Question about β 1 hat and βº hat, the last formula introduced show you should use some cov, var and mean. I know you can have those related to the sample and/or the total population . Which one should be used ?
Hi your videos are very helpful but what is covariance and variance because I watched your first video and nothing explains that. Thanks!
very helpful esp for me, which the beginner to learn econometrics, i have one question regarding to “error term” so when we applied this formula, we just have known what the error term is? or i mean thet the error term is decided by us or not? i just still confuse about the error term..
love you !!!
great lecture more please
Hello I have a question. I want numbers and analysis in my working life. Does choosing economics and business economics, rather than econometrics and operational research for undergraduate study, affect the carreer paths I will have? The thing I am trying to ask is can one do the other's work without that much additional work? I would be very if you could help me.
thank you it help ed me alot
a very helpfullllllllll... thanks
It helps a lot
thank you sososososososo much!!
Amazing.
Shouldn’t the equation be: ûi = Ŷi - Yi
@starostadavid
4 жыл бұрын
Nah, measured data got to be the core value. It doesn´t matter if u get a negative value, because you sqare it anyways.
@LegionFan
3 жыл бұрын
@@starostadavid to me it also seemed that having Yi first allows you to see if the actual data point is below (negative) or above (positive) the linear regression estimation.
12.24 - 30 The mean value of y (Laughing) has anyone noticed that?
@nurkardinamassijaya2107
7 жыл бұрын
Yes I recognized that lol
@Usas12fann
6 жыл бұрын
Laughing about the magical simplicity after 12 steps of derivation I guess haha.
niceeee video sir..thxxxx
Hai Chris, in this video , you have mentioned about the causal relationship.You were saying that 'we cant assume that there is a causal relationship between wage and education. can you help me to explain in more details on that..? i kinda didnt get what you have said.must we assume?? pls help..:(
Based on your explanation on the graph, U^=Y^i - Yi not Yi-Y^i @ 9:50
@margaritaosadcha7031
9 жыл бұрын
HL Lee yes, I am was also confused about that. So its Y^i - Yi then? is that right?
@littlebanana7372
9 жыл бұрын
HL Lee i guess because over the line are positive areas.. and contrastly...
@ashy99591
7 жыл бұрын
negative error term. he meant
very helpful thanks
Please post more videos
Unfortunately, the notation depends on the textbook and a lot of the notations overlap. The "regression sum of squares" (RSS) is the equivalent of the "explained sum of squares" (denoted ESS or SSE). However, be careful because the sum of squared residuals is usually denoted as SSR, but it is sometimes referred to as the RSS (residual sum of squares). Lecture 3 starts to explain what the SSE (explained or regression sum of squares) is -- In short, it is the amount of variation in y_hat.
13:19 wow that car just goes Zoooorrrm
oh my god I kept screaming Eureka in my head, everything is so clear when he explains it
Are there assigned textbooks and materials for this course? E.g., De La Fuente, Wooldridge, etc.?
Thank you so much
IT IS NICE LECTURE
I don't understand the "betas" zero and one about covariance and variance.
@alieverbol
5 жыл бұрын
beta zero is intercept and beta one is a slope
i think you forgot the link to video 1 in the corner. video 1 was great!
How do I get the quiz for practice?
please make more!!!
I can't seem to understand this equation "ui = yi - yi_hat = yi - (beta0_hat) - (beta1_hat)(xi)" If you look at the line drawn up the yi_hat have a longer distance to the regression line than yi. So, in my head, to calculate the ui you should turn that equation over like "yi_hat - yi"?
@aaaaa8744
7 жыл бұрын
I agree
I know you haven't uploaded in a while but please put up a 4 minute long loop of the music at the end. I really want it on my mp3 player. Cheers.
thanks a lot
Bro wtf this is great
Just wondering regarding the Y estimator equation. Y hat= B1 hat + b2 hat Xi + U hat Y hat should be changed to Y or keep Y hat and remove U hat? Since the estimator of Y refers to the fitted line rather than sample distribution
@stijnlijnsvelt5166
3 жыл бұрын
Did you find out the answer? I was asking myself the same thing.
can you use B1 instead of B0
Hey, this is Chris from Keynes Academy! Have you enrolled in an econometrics class? Join your classmates and stay ahead of the curve by subscribing! All questions and feedback are welcome -- Please comment, message or e-mail us your thoughts!
Why did you stop making the video lecturers ?
SSR is the sum of squared residuals. I understand what this is, the distance between the data point and the best fitting line squared. In my course RSS is the regression sum of squares and I don't quite understand what that is.
Additionally, if you want to estimate any ui, find the estimates for the intercept and slope and plug them into the ui equation: ui = yi - yi_hat = yi - (beta0_hat) - (beta1_hat)(xi).
Interesting
Hi, can someone please explain to me like I'm 5 what a "TRUE MODEL" is? I still cannot comprehend why we have a true model when we are exactly just estimating things? For example, if I want to find the relationship between wage and education, where is the TRUE MODEL from?
at 9.30 surely you mean Ui(hat) is equal to Yi(hat) subtract Yi? not Yi subtract Yi(hat)
@BIGBEAUTIFUL22
9 жыл бұрын
NavHDpoop i was wondering about that as well
I'm sorry but can anybody tell me what is the meaning of using HAT? what's the difference?
If Y^ = B0 + B1*X +U^, then why do you need to remove U^ to get Y?
@stijnlijnsvelt5166
3 жыл бұрын
first this equation is said and then to get Y^ we have to do -U^. That does not make any sense to me. Could someone explain?
Hey, mr Keynes! I really enjoy your videos! Will you be posting new ones during this summer? I'm thinking of enrolling for an econometrics course during the fall. Could you recommend any good literature? Many thanks in advance! //Ali
Bata 0 is when the value is 0 or where the line inter the y when x =0
Thanksssssssss
Can you pleaseee reply to the question about your subtraction of Yi hat from Yi
@taijaskumar5331
9 жыл бұрын
BIGBEAUTIFUL22 All residuals below the line are -ve and all residuals above the line are +ve. therefore, below the line --> Ui(hat) = - ( -Yi + Yi(hat)) = Yi - Yi(hat) or it could be , - Ui(hat) = -Yi + Yi(hat) Hope that helps. Cheers
good lecture. but you need to add some clarification using while you use formula.
thumbs up
Hi! I think that these videos are very helpful. Regarding the simple regression model I think there is a mistake in the first expression of ^Y, which should be ^Y=^Beta_0+^Beta_1 X instead of ^Y=^Beta_0+^Beta_1 X+û. However, the expression is correct when it is derived from the graphical representation.
@xorenpetrosyan2879
9 жыл бұрын
no it shoulden't,
I think his formula is correct..I checked my notes and have similar notation Yi-YiHat
Where's the quizz??
lol the amount of notation you see from one school to the next, let alone by my own school. Each year i need to learn new notations to replace previous ones
get back on your grind and post more videos
Shouldn't it be Y^ - Y instead of Y - Y^ because U^ + Y = Y^
how is. Ui hat = Yi - Yi hat if , Ui hat + Yi = Yi hat
I want to take the quiz =-=
2020 covid 19.... we here