Back to Regression Models

4.4

2,562 ratings

•

439 reviews

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....

Mar 13, 2018

Great course, very informative, with lots of valuable information and examples. Prof. Caffo and his team did a very good job in my opinion. I've found very useful the course material shared on github.

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.

Filter by:

By Roman

•Mar 10, 2019

Really bad. Worst of the whole Data Science spezialization. Bored me to death. Lecture had nothing to do with the quizzes, quizzes had nothing to do with the final assignment, final assignment had nothing to do with the lectures. Fight through it, there is light at the end of the tunnel.

By George C

•Apr 30, 2018

This is the worst course in the series. Caffo does a terrible job at explaining regression, the final assignment requirements aren't properly addressed, and it appears they didn't quite spend time on how to make it all work (2 pages to test out different regression models, make an inference, and everything else is absurd). I highly recommend avoid this course, and instead go through the R guide on linear regression; in the end, I used those to get through this course.

By ALEXEY P

•Nov 18, 2017

If you thought that the previous course (Statistical Inference) with Brian Caffo was a horrible experience -- think twice and get ready for Regression Models. It is way worse. Imagine an instructor starting his explanation by showing you some (rather involved) formula and immediately jumping to the discussion of the various terms without actually telling you clearly what this formula is for and how to use it. Then you will get a pretty good idea about the instructor for this course. He is a horrible teacher, who clearly does not understand what teaching is and how it should be done properly. Total waste of time.

By Claudio F S

•Mar 11, 2019

This module was the maximum. I learned how powerful the use of Regression Models techniques in Data Science analysis is. I thank Professor Brian Caffo for sharing his knowledge with us. Thank you!

By Johnny C

•Sep 25, 2018

One video is wrongly edited, half of it is repeated. The instructor gives too much information and is difficult to follow, some information is even trivial.

By Ricardo M

•Jan 30, 2018

Just like the previous course in the specialization path (Statistical Inference) the course delves into some relevant topics however it doesn't feel as properly structured. While on the first week the lectures seem to try to give a basic and comprehensible learning of linear regression, once we start into the more advanced topics it gets confusing.

Lots of formulas and concepts thrown at you without much clarification. For someone without any knowledge/background on statistics this can be quite difficult to grasp the concepts.

The module for Poisson Regression is very poor in terms of information. just feels like a very light overview of the matter.

The course should be reviewed or at least the indication of "Beginner Specialization.No prior experience required." should be updated to mention that some knowledge in statistics is recommended .

By Joana P

•Jan 26, 2018

Honestly the materials of this course are really confusing. So many focus on the mathematical value instead of real examples and scenarios to use the concepts reached. Also it would benefit if there was a clear message coming through, like Machine Learning course where things follow a order.

If it was not by the book of Mr.Field with Statistics in r, I would never be able to understand what was really being said in this course. Or what was the best strategy to effectively do a proper regression analysis and what would be the best models.

By Nicholas A

•Dec 22, 2017

Personally, I am not a fan of this professor. He over-explains all of the topics, just to tell you at the end of the lecture that you don't need to know the specifics and can do it all with one function. He is very unengaging, difficult to follow, and rushes through lectures. And finally, HE BLOCKS THE SLIDES WITH HIS HEAD SO YOU CAN'T SEE THE NOTES. I feel like out of all the professors in this specialization course, there were so many others who could have taught the material better, especially since this is probably the most important course of the entire specialization. I feel like I only began to understand the material once I finished the course project, and even then I have no idea how regression models work.

I'm now going to be taking a month or 2 off from the courses to read more about statistical inference and regression models on my own, since I feel completely unprepared for the upcoming Machine Learning course.

By cleoag1

•Oct 29, 2017

Very math heavy and not super useful for psychology students.

Without a tutor, that I had to pay $30 an hour in addition to this course, I would not have passed.

The layout was rather convoluted, there were several spelling mistakes (one that completely changed the meaning of a QUIZ question) and it was not as conceptual as I was hoping for.

The conceptual limitation is big for me as I don't care about the math, I'm a psych undergrad trying to learn statistics for my honors thesis, not a math course.

It also made it difficult to apply what we learned since the data we worked with wasn't that easy to understand and was incredibly boring (car mpg data and insect sprays??).

I'm also slightly upset that coursera signed me up for a subscription when all I wanted was one course, very cheeky.

By Jeffrey G

•Oct 18, 2017

I was optimistic about this class because it started out fixing some of the pedagogical mistakes the professor made in Statistical Inference, but by the time we got to week 3, it was pretty clear that the course was trying to accomplish too much in 4 weeks, and instead of focusing on the most important parts of regression and making sure they were taught well and understood clearly, I feel the course tried to do far too much. The only reason I gave it two stars instead of one star was the course project was relatable - choosing the best transmission for maximizing mpg is a real-world problem that I can (and did) have a discussion with my mother about. Too many assignments are about something completely inane, like guinea pig teeth or flower petals. If you're going to inspire students to learn the material, the examples (and data) must be relatable to them.

By Ravi K

•Jun 12, 2019

This is really a nice training

By Alán G B

•May 28, 2019

It is an excellent initial approach to Regression Models. I was able to apply some of the models in my work. Further analysis of the mathematical and statistical theory is highly recommended.

By YANAN D

•May 24, 2019

Really Helpful

By Nino P

•May 24, 2019

Similarly to statistical inference, this is a bit harder course in the specialization. Still passable and recommendable.

By Andrew

•May 16, 2019

Great introduction to regression models. A ton packed into the class. Be ready to be challenged, but you'll learn a lot.

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 Ekaterina S

•May 12, 2019

It was a very usefull course. It is a very good approach to the theme - the main essence without much math difficulty.

By Diego C

•May 04, 2019

Very good course. Though basic, it provides you with the first tools and knowledge. The forums aren't what they used to be it seems, but you can find almost any answer there from past courses.

By MEKIE Y R K

•May 02, 2019

Really interesting and full of advices.

But would like to dig more into the Logistic and poisson regression residuals explanations :)

By Rodrigo O

•Apr 16, 2019

great

By Don M

•Apr 10, 2019

Overall an excellent course, but there were some issues with the wrong function being specified in one quiz (Q3q6) and the wrong answer in another. Apparently it has been that way for years, according to the forum. The quality of the lectures was very high and the information interesting, so compliments to Dr. Brian Caffo on that. However, the estimated time for completion of each week is ridiculously short compared to reality. Five hours? For me it was more like 20 hours, and more if I did all the Swirl exercises. Such low-balling on the time estimates is typical of the Data Science stream. The final project is given as 2 hours but it was closer to 15 for me. i wish Coursera would go back to the stream model where you could bump yourself to the next intake. That is much less stressful for busy working people like me.

By Satish V

•Apr 08, 2019

The instructor's delivery and content, although very professorial was very dry. For students who don't have that much of a background in regression and statistical inference, I think it would be good to get to the gist/summary - i.e the what (what kind of problem we are trying to solve) and the how (how to do it in R and more importantly how to interpret the results).

By Dora M

•Mar 30, 2019

Good class.

By Yadder A G

•Mar 28, 2019

The course was incredible. You can learn a lot of skills about regression models and even more. It would be incredible if the course could have more examples or little excercises.

By Manny R

•Mar 22, 2019

Really Fun Course. There is a lot to learn in this topic and this could be studied for a lifetime. I feel like I could apply this to discover solutions for issues at work.

Coursera provides universal access to the world’s best education,
partnering with top universities and organizations to offer courses online.