In this course, we will expand our exploration of statistical inference techniques by focusing on the science and art of fitting statistical models to data. We will build on the concepts presented in the Statistical Inference course (Course 2) to emphasize the importance of connecting research questions to our data analysis methods. We will also focus on various modeling objectives, including making inference about relationships between variables and generating predictions for future observations. This course will introduce and explore various statistical modeling techniques, including linear regression, logistic regression, generalized linear models, hierarchical and mixed effects (or multilevel) models, and Bayesian inference techniques. All techniques will be illustrated using a variety of real data sets, and the course will emphasize different modeling approaches for different types of data sets, depending on the study design underlying the data (referring back to Course 1, Understanding and Visualizing Data with Python). During these lab-based sessions, learners will work through tutorials focusing on specific case studies to help solidify the week’s statistical concepts, which will include further deep dives into Python libraries including Statsmodels, Pandas, and Seaborn. This course utilizes the Jupyter Notebook environment within Coursera.
Week 3 starts to get unreasonably difficult and hard to understand. Apart from that, the course is still worthwhile to take.
I think the notebook walkthroughs, while useful, could use some extra reinforcement in the statistical concepts
Great course. It really improved my understanding of statistical modeling methodologies.
Great course. In my view, the lectures were too long and the assignments a bit easy. But, overall, great course.
Good course, but the last of three was the most difficult one. I hope that it were a good introduction to the fascinating world of statistics and data science
The specialization covers important practical topics. I am glad to have the opportunity to explore it.
Very informative. But had few confusions in the last course. Also the python code explanations were not good as the instructor was rushing through it without explaining.
Just like the other courses in the specialization, very well thought out and planned! Up to date, great professors . . . couldn't ask for more!
Good course giving a fair view on fitting statistical models. Could do to elaborate on some of the theoretical models using more illustrations for more understanding.
Challenging but excellent course, especially how content was organized and examples used to explain concepts
Overall it's very good for someone who has a fair background in statistics, except for some small mistakes in slides and notebooks.
These whole three certifications lays the foundation for learning Machine Learning a more in-depth way.
Showing: 20 of 143
If you don't already understand the topic don't bother with this course, the lectures are 95% hand waving and showing formulas they don't explain how to make sense of and then the quizzes are answering questions on what they didn't bother to explain.
I was looking for an application course that would help with using Python with real world data. This was a theory course that added a small poorly explained notebook and a very brief lecture which didn't explain the code very well. If you're looking for a statistics theory course this might be fore you. If you're looking for how to use Python in the real world, I might look at other courses first.
Great lecture content, poor quiz design. Hard to apply any of the concepts that you learn.
Overall this course was okay at best. It DOES NOT lack depth nor are the notebooks poorly explained. Many high level mathematical concepts are covered in this course and it is not shallow at all. The python notebooks are robust, and are excellent examples of statistical coding. But it badly lacks a bridge to take the student from simple theory to high level theory, the lectures are very poorly designed and are just bad at transmitting the subject content, critical explanations of terms and mathematical processes are lacking, and I had to google many intermediate statistical concepts and explanations just to understand what was going on; this is not a course for people with no statistical and probability background
I was really disappointed with week 3 and 4 of this course and only managed to learn a few basic lessons despite being able to pass the quizzes. I would recommend that they take out course 3 of the specialization and only add it back after revising and revamping course 3.
17/3/21
The most impressive part is Week 2 Linear and Logistic Regression model fitting, Professor Brenda is Brilliant! She has the magic to explain complicated and abstract concept into a very easily understandable ones. Thanks her a lot! Also I was impressive on Week 4 Bayesian approaches courses. Thanks Mark Kurzeja. I think He is a very qualified teacher and prepare for this course content very careful and take it seriously. He also gives a very clear mind to understand those abstract statistic concept!
Overall, the series of Statistic with Python are impressive! You can really learn something useful and the course design is scientific. All teachers in all courses are very good!
Well-structured and adeptly delivered course.
A perfect introduction to regression analyses and more advanced statistical modeling procedures that are frequently used in practical scenarios to conduct in-depth data analyses and make accurate data-driven predictions. Students, independent learners and industry professionals who wish to understand the intricacies of assessing good predictive models can start off their analytical journey with this course.
I think the content here is great and Mr. West is a wonderful teacher. That being said I do believe the multi-level regression model topics were quite difficult to understand and it did feel like some of the content was a bit rushed in week 3. It would have been nice to go over some non-linear regression as well as I did appreciate week 4 but I am not sure these special topics were as useful as the previous topics. Weeks 3 and 4 could have been used to dilute the content a little bit to go into multi-level regression a little bit more in depth and maybe look at non-linear regression. On a side note, I though the pdf files explaining linear regression and logistic regression as extra reading were absolutely fantastic to clear things up. I am sure the course would benefit from more content like that.
Like the other courses in this specialization, way too much theory covered, and the easy quizzes and labs give the learner a false confidence that he/she's mastering statistics. Instead, you grasp some of the theoretical knowledge, but not of the underlying math and therefore none of the intuition. The same is true of Python, all that's required is to hit the run cell button, no actual coding is required.
The lecturers are super enthusiastic though, and the final week was fantastic. Mark Kurzeja should have his own course on probability and Bayesian statistics.
Week 3 of every course has been super dense, and I think T Brady West should have his own course on sample design and weights because right now his lecturers drag down the overall quality of the course. It's all slides and text, math is brushed over and not enough of it is applied. Honestly, if you wanted to really get into Multilevel & Marginal Models you'd need 4 weeks.
My advice, take the AP statistics course on Khan academy, watch some STATSQUEST on youtube & perhaps take the intro to statistics offered by Stanford University. You can also take this course/specialization and just skip weeks 3. You can probably pass the tests anyway
Here's my rating by week.
Week 1: 4*
Week 2: 4*
Week 3: 1*
Week 4: 5*
The course inspires me to think more about how to use statistical theory in some application fields. Specially the python exercises such as multilevel regression and marginal regression is very helpful in understanding their concepts. However, in my opinion, it is better to add some more pratical Python source code or give some learning links in Github. For example, I have not understanded what is Bayesian regression even with the help of the source code given by the course until I found a source code in Github. From my own experience, a good piece of source code surpasses a long-time oral explaination.
Thank you for creating this course. I have learned basic knowledge to succeed my incoming business education. I have a bachelor degree of laws and am transferring to a master of management. I used this course to learn the prior knowledge that I need about statistics. I finished this specialization and feel more confident about the numerical analysis. Thank you again Michigan Online for your great courses!
I enjoy the lectures. The professor has a good speaking and teaching style which keeps me interested. Lots of concrete math examples which make it easier to understand. Very good slides which are well formulated and easy to understand
Really well explained, maybe a bit long on some aspects but really great overall. The best of the three courses especially considering that's the "practical one"
Good course, but the last of three was the most difficult one. I hope that it were a good introduction to the fascinating world of statistics and data science
Awesome overview about what can we do with statictics knowlegde! Half theory, half practice with Python is a great format
Challenging but excellent course, especially how content was organized and examples used to explain concepts
Really thorough and in-depth material about statistical models with python.
This course does a nice work introducing the concepts of model fitting, especially during the first two weeks where the emphasis is on multiple linear regression and logistic regression. Professor West does a great job focusing on the theory that one needs to know before applying any modeling, and there is quite a lot of Python material at the end that the learner will have to explore mostly on his own, since the corresponding videos are somewhat lacking in depth. Week 3, on the other hand, introduces some very interesting but advanced concepts that can be quite hard to grasp, especially for learners that haven't had much experience with classic statistical model fitting. Week 4 is mostly an introduction to Bayesian Models, but nothing deep.
Overall, I was a bit disappointed with how the course was structured, and the fast pacing after Week 2 might discourage learners. I would recommend the couse however to anyone wanting to really follow up on the material covered, especially from a Statistics perspective (not Data Science-wise).
Overall a fair course , but i felt it was a bit too fast paced and more focused on theoritical statistics with serious lack in Pyrhon practising.I mena the notebooks were a great deal but the instractions on them and the video coures were not what i expected compared to previous lectures.It was a little bit difficult to follow on with the theoritical courses - weren't explanatory enough for me. And for sure i needed more Python practsing , lecturing and of course assessments.
Starting from week 3, the lecture becomes very confusing with a lot of unexplained terminologies. Very few examples were given to help understanding the concepts.
In this "Fitting Statistical Models to Data with Python" course, you will learn about 1. Variables, multicollinearity, study designs 2. Fitting statistical models to independent data - continuous (linear regression) and binary dependent variables (logistic regression) 3. Fitting statistical models to dependent data - Multi-level models (model fixed and random effects on the intercept and cluster terms) and Marginal models 4. Introduction to the Bayesian Statistics The course covered lecture videos, well-prepared readings, Jupyter notebooks to introduce concepts as well as practice notebooks, lab walkthroughs and quizzes. Brady may speak alittle too fast, especially when it comes to long sentences, so you may need to rewind certain segments of the videos numerous times to revisit some concepts as you reflect and learn. Discussion forums were also actively monitored by a TA who got back to me usually within a day, which helped to unblock conceptual roadblocks quickly. You may encounter some issues with the Coursera platform: 1. Labs may fail to load at times, even after following instructions to restart the machine which your lab runs from 2. Your discussion forum comments may disappear right after you post. Remember to copy your comment somewhere else (e.g. on a notepad) as you may need to refresh the page and post your comment again.