6. Regression Analysis

MIT 18.S096 Topics in Mathematics with Applications in Finance, Fall 2013
View the complete course: ocw.mit.edu/18-S096F13
Instructor: Peter Kempthorne
This lecture introduces the mathematical and statistical foundations of regression analysis, particularly linear regression.
License: Creative Commons BY-NC-SA
More information at ocw.mit.edu/terms
More courses at ocw.mit.edu

[@zhuangjiwang5392]

This lecture, compared with the previous ones, provides a great example for the importance of using blackboard in math courses.

[@user-jx2lj1nd6s]

Lol that's true

[@datafabrics5039]

the old fart is allergic to chalk

[@SeikoVanPaath]

Timestamps: 0:02:40 Overview 0:29:10 Ordinary Least Squares (OLS) Estimates 0:45:54 Gauss-Markov Theorem 0:54:47 Generalized Least Squares (GLS) Estimates 0:58:17 Normal Regression Models 1:19:25 Maximum Likelihood Estimation

[@connordavis4766]

thanks!

[@phillustrator]

This lecture convinced me that no amount of money you pay will guarantee you good teachers.

[@burnbush]

Awesome content at least my knowledge of matrices and statistics has been broadened on a larger scope. Thanks.

[@user-ok4wr4zm5i]

Excellent lecture

[@liuauto]

The slides ruined the lectures. This is a math course. Writing on the board can save more time than talking and hand waving

[@olzt100]

That's a major differrence between a high end college and high school. You have to figure out how to make the system work for you and not vice versa. Students will find most of the world operates that way.

[@bishalbhattarai0425]

jie liu n

[@lessmoneylessproblems5145]

Holy, who knew that linear regression could get so complex.

[@fustilarian1]

I miss Choongbum. This guy shouldn't be allowed to teach -.-

[@caverac]

@30:30 There's a mistake in the slides: the last term in the sum should be \beta_{p}, not \beta_{i,p}

[@johnhart1790]

There is also a mistake in the matrix X, the last entry should be X_(n,p) not X_(p,n).

[@aboubacaralaindioubate6086]

It's a '"typing-mismatch". ( Erreur de frappe, in French). Not a logical mistake. : )

[@azizlarabi1941]

@@aboubacaralaindioubate6086 In english 'typo'

[@user-bt7u12gh5]

52:07 another typo? shouldn’t it be E[f’y]=f’E[y]?

[@SergeiIakhnin]

Good content, but should have really been handled in two lectures. A lot of time is spent on the basics, and then all of the more advanced details are simply glossed over due to a lack of time.

[@anant088]

No explanation post GLM!

[@KARAB1NAS]

This a good example of how a course in maths should not be. Especially the notation is so dodgy - Random matrix X is referred to the matrix of realizations of X all the time. Ridiculous

[@connordavis4766]

In 10 years of math education and a PhD, I have never seen a talk that was improved by having slides including my own dissertation.

[@thalberg-]

@@connordavis4766 why are mathematicians against using slides?

[@connordavis4766]

@@thalberg- Math slides tend to be overly full of text with lengthy theorem statements and calculations. In principle you could cut down on these but in most cases those long list of assumptions or the mechanics of a calculation are *the whole point*. Slides are appropriate when you're not trying to go into any detail at all and you just want to give a surface level overview of something.

[@gamebm]

Slides may play a part, but IMHO the main reason is the content itself. I guess most ppl have some background in the previous lectures by Lee, so that was nice and easy. The content of this lecture is new to most ppl, it includes more stuff, and many details are somewhat skipped. Even one tries to derive everything on the board, they would have to go at a faster pace, and as a result, most ppl still won't catch up easily.

[@gustavobolssonbilibio370]

arggggg he knows how to read slides.

[@roymarshall_]

This was almost every single class I took in college. Good to know MIT isn't really any different.

[@joem8251]

uh...

[@chunlangong2214]

All his slides are in the textbook. Why the students in the classroom?

[@kalinda619]

Welcome to lectures by researchers being forced to teach 😭😭🤣

[@lochestnut]

can anyone help me prove that: if \epsilon ~ N_n(0_n, \sigma^2 \Sigma), then \Sigma^(-0.5) \epsilon ~ N_n(0_n, \sigma^2 I_n)

[@holidy1]

Not sure about the equation. Maybe variance(a*x) = a^2*variance(x) helps.

[@dankole307]

With regression analy. The order of the fit justifies weighting. Seems to me neural networks are a much better subject to fits. Both are worthwhile. Neural nets provide options for rapid changes and simulation.

[@CaseyVanBuren]

What if I told you the theory is the same, just several layers of linear models.

[@WallaceRoseVincent]

Anyone interested in working through the course together?

[@VasuDev-kg6uq]

I am

[@WallaceRoseVincent]

@@VasuDev-kg6uq Where are you from?

[@VasuDev-kg6uq]

@@WallaceRoseVincent India

[@VasuDev-kg6uq]

@@WallaceRoseVincent check your inbox

[@miladresketi7392]

I have created a discord channel for this course, you can join If you're interested discord.gg/A2myKzU

[@danielduranloosli]

That was a brutally dense lecture with almost no real-life analogies. At that pace, you would need to be a linear algebra god to actually have the time to think about the statistical interpretations and applications of the expressions you are following. Also, no use of much-needed graphs or technology whatsoever.

[@toshb1384]

it’s really not that bad lol

[@dennisestenson7820]

Good thing you didn't have to pay to see this lecture.

[@TroubleMakery]

This is not a course on linear regression, it’s just supposed to be a revision on the topics in finance. This is merely supposed to be a refresher so the more advanced stuff is more easily understandable. That’s why I’m here.

[@raneena5079]

No lol, all of the linear algebra was very basic

[@forheuristiclifeksh7836]

5:57

[@xu6845]

I really can't follow that

[@michalroesler]

what it means and why it's there (y - theoretical y)^t * (y - theoretical y). why ^t is there?? what it means? kzbin.info/www/bejne/omLOfXaorbFsfs0

[@raneena5079]

this is just the square of the Euclidian norm (basically size) of y - theoretical y. y - theoretical y is the error, so if you minimize that quantity, you are minimizing the error of your model.

[@user-ok4wr4zm5i]

The previous lecturer did not cope with his work

[@CaseyVanBuren]

These slides seem riddled with mistakes, indices are lost or flipped or added where they shouldn't be. I was expecting more from MIT

[@amandinelevecq6664]

Now I remember why I didn't like statistics and chose maths at university 😄

[@phillustrator]

I don't know why, but the storytelling of statistics is almost always terrible. I start to dose off every time I try watching a lecture. Including this one. It doesn't help that stats teachers often have the charisma of an accountant on Xanax. (yes, this one included)

[@pramesh.gurung]

isnt stats part of math??

[@_danila5185]

@@pramesh.gurung it is

[@ArnobAlam]

Is that dude using a Mac at MIT?

[@zhulin85]

seems not. looks like only a dock app or theme.

[@harshd1122]

You are calling him "Dude"? Really? He is teaching at MIT. Show some respect. His name is Dr. Peter Kempthorne.

[@tyqwe45qe]

@@harshd1122 He's talking about the student...plus, even if was about the professor, who cares. Stop being so sensitive

[@power9k470]

​@@harshd1122 The good old deference to authority.

[@power9k470]

​@@harshd1122 The good old deference to authority.

[@Diana-yl1jo]

一脸懵逼 这几节课真的都太概括性的总结内容太多了 每一项扒出来都得消化一阵子。。

[@milesreynolds6128]

实际上不如每一节课的内容都去单独学一个，这个每个标题都讲个皮毛和没讲也没啥区别

[@ehsan106]

This is a disaster . I think MIT should remove all the videos and never allow these people waste our time again.

[@WaveWatchTrading007]

This man talks in too many abstractions.