Regression Models に戻る

4.4

2,729件の評価

•

463件のレビュー

Linear models, as their name implies, relates an outcome to a set of predictors of interest using linear assumptions. Regression models, a subset of linear models, are the most important statistical analysis tool in a data scientist’s toolkit. This course covers regression analysis, least squares and inference using regression models. Special cases of the regression model, ANOVA and ANCOVA will be covered as well. Analysis of residuals and variability will be investigated. The course will cover modern thinking on model selection and novel uses of regression models including scatterplot smoothing....

Dec 17, 2017

Excellent course that is jam-packed with useful material! It is quite challenging and gives a thorough grounding in how to approach the process of selecting a linear regression model for a data set.

Feb 01, 2017

It really helped me to have a better understanding of these Regression Models. However, I've noticed that there is a video recording repeated: Week 3, Model Selection. Part 3 is included in Part 2.

フィルター：

by João R

•Aug 20, 2017

Needs more practical examples. Could be rerecorded. I love mathematical theory but past week 2 it is really too theoretical, in my opinion.

by Barry S

•Mar 15, 2016

This course is the first one in the Data Science series to lapse in terms of the clarity of the lectures, and the sense of cohesiveness of the material. Brian Caffo's lectures in Statistical Inference were good; in this course they seem to veer left and right rather than get straight to the essence of whatever subject he is lecturing about.

A more structured final project would have been helpful. The instructions on this project weren't quite so blunt as to say "Take this data set, do some regression-y stuff and come back with something about these two variables," but that's basically as far as our instructions went. It could have been a great learning experience to have a more detailed guide through the construction of a regression analysis, but instead an assignment which was 40% of our grade was put together as an afterthought. It was the assignment equivalent of stopping in the 7-11 a block away from a birthday party to buy a card.

Also, in terms of delivering the content: Mr. Caffo needs to structure his slide/video arrangements so that he is not standing in front of the text. Think of it from the point of view of somebody wanting to listen and read at the same time.

by Brian S C

•Mar 01, 2016

Overall okay course but the lectures are too focused on theory with some applications to the real world. I think this course needs to be reconfigured and taught from an applied focus instead of 30% applied 70% theory.

Also the new format is horrible and TAs are nonexistent as are discussions in general on the forums now. The TAs were a critical learning component before especially considering that unlike on EdX where course staff actually participates in the forums, on Coursera I do not think I have ever observed course staff actively participating in the forums.

by Brian

•Feb 12, 2016

way to much emphasis on non-data science. This one course covers more information that the rest of the courses combined..

by Ritu B

•Feb 07, 2016

Appears more like a revision for those who already know the content than geared towards those new to the subject.

by Thej K R

•May 13, 2019

Worst teaching by Brian Caffo! typos in quizes after 4 years even. And brian has put very littel effort into making it digestable for students. Look at his lectures on youtube and I have commented at each lecture! So bad. A simple googling outside of his notes was so much more better for understanding regression!

by Shahryar N

•Nov 29, 2018

The course is awfully simplified.

by louis d

•Jun 11, 2016

Content and quizz are not aligned.

Mentors answer to 0% of the forum posts.

Poor student community.

Do not pay for this course, just follow the swirl and/or get some tuto about regressions.

by Martin L

•Jul 26, 2017

Very poor - the worst of the specialization courses by far. The lectures are confusing and poorly presented. If you want to understand regression you'll have to look elsewhere.

by Eric T

•Feb 21, 2017

Important material, poorly taught.

by Robert O

•Apr 06, 2016

Very little depth. I don't recommend this if you don't already have background in statistics or R. I really didn't learn anything. I mostly just gamed the quizzes and projects.

by Satakarni B

•Nov 08, 2016

Not worth it for the bucks.

Instructor has tried his best to make no sense of the subject.

I would be happy for a refund.

by Fabiana G

•Aug 31, 2016

I was really disappointed with this course. I took the other courses from Brian Caffo and truly enjoyed them. For the previous courses, I've always used the books and they helped me tremendously to be able to comprehend the material. There is a book for Regression Models but but it's a real mess. It feels like a draft that no one cared to take a second look. There is a bunch of wrong code and typos. The explanation doesn't go as far as it should. I had to resort to many different sources just to be able to get by the course. I hope the instructors review this course soon because it does not have the same quality as others. If they don't review it, don't bother paying for it. Try learning Regression Models elsewhere.

by Vineet J

•Jul 30, 2017

not good tutor

by Stephen E

•Jun 27, 2016

To be honest I don't think this is worth the money.

by Derek P

•Aug 18, 2016

The course is essentially just a review of formulas with very little intuition explained to the beginner. It was necessary to use a collection of outside material from other courses and readings to learn the concepts. This course needs to be completely redone with a focus on developing a student's intuition for the material and then support this intuition with basic examples that build as the course progresses. A fundamental demonstration of how to use R to work through regression models (starting from square one) should be added so that this becomes a self-contained course. As it currently stands it is a collection of poorly integrated slides and concepts that serve to confuse the student more than educate. Other classes teach this material infinitely better.

by Matt G

•Feb 15, 2016

Poorly designed, executed and instructed. Too much is left off the materials.

by Tom

•Jul 22, 2017

Terrible. If you want to learn about regression, even in R, go elsewhere. This course damages the brands of Johns Hopkins and Coursera...anybody heard of quality control?

by Adnan B

•Aug 13, 2019

This is the whole course that kind of discouraged me persuing data science field... i wish i wish i wish there was different instructor