Comentários e feedback de alunos de Fitting Statistical Models to Data with Python da instituição Universidade de Michigan

Another interesting course - the final one in this specialisation - but the difficulty really ramped up in Week 3 after the final peer marked assignment. I had been so impressed with the clear explanations, revision and review, and the opportunities to apply new knowledge. However, it all became very abstract - I thought Mark did a good job but perhaps Bayesian is a whole different specialisation. Overall, I really enjoyed the specialisation and I am pleased to have received a good grounding in statistics ahead of my Data Science diploma. Thank you to Brenda and Brady especially but everyone was very strong and the future is bright with some enthusiastic young talent coming through at Michigan. Edward

It is a good introductory course for statistics. The programming assignments were not challenging enough to cement what you have learned. The concepts in week 3 and week 4 were challenging and their approach was not good. I feel like I wasted my time. The focus should have been on multilevel model fitting rather than covering bayesian statistics. Week 4 only added more confusion. However, as an introduction course, they did a good job of presenting the concepts in the prior courses of the specialization.

The videos were really lengthy, above 15 minutes videos are hard to understand for me. Although the overall specialization is really good and gives me very confidence. I would recommend everyone who wants to be a data scientist in future.Thanks Brenda and Brady T West and of course Julie Deeke and other students.

The course is alright. They give a high-level overview of linear and logistic regression, and dip a little into Bayesian statistics.

Note that they use the StatsModel package in their practice assignments. So I was a bit disappointed I didn't get to practice sklearn, which is about x10 as popular in the field.

python codes were pretty tough to undertsand in the end but the concepts though difficult to understand the faculty did there best possible to make it understand. Python codes should have got little bit more time to be explained

Overall, the course was a great refresher of statistical theory and application with some great Python exercises. However, some of the Python coding instruction itself could have been more detailed.

A great introduction to regression and bayesian analysis in python. I get that the content is hard, but they sum it all well. I would recommend for those who have prior knowledge of statistics.

The code examples may be more precise with detailed comments. Some codes are not understood, in other words codes can be refactored in a way that can be more suitable for reproducible studies.

Very good course, I like many practices and evaluations focused on database of real cases, perhaps it would be advisable to reproduce results from the same sources .....

JL

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.

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.

Good for advance topics like Marginal and Multilevel modelling. The Bayesian model could be explained in a detailed manner by providing more python assignments.

The course is great, the only improvement I would make is to be a little more didactic in the last two units because it is a more complicated subject.

The course was wonderful however, sometimes I felt that a little bit more details could be provided when python code was being explained for week 2.

Overall it's very good for someone who has a fair background in statistics, except for some small mistakes in slides and notebooks.

Me gusto sobre todo los modelos de nivel combinados con estadistica bayesiana ,eso fue lo mejor y de verdad invaluable del curso

Week 3 starts to get unreasonably difficult and hard to understand. Apart from that, the course is still worthwhile to take.

Great course. In my view, the lectures were too long and the assignments a bit easy. But, overall, great course.

I think the notebook walkthroughs, while useful, could use some extra reinforcement in the statistical concepts

The course is great, but I would suggest that the subject of week 3 be divided into two weeks.

Great course. It really improved my understanding of statistical modeling methodologies.

team work

None

The content of this course is very thorough, but unfortunately it does not make very good use of the online asynchronous nature of a platform like Coursera. Most of the course consists of lengthy video-lectures paging through slides (and occasionally walking through notebooks). The hands-on parts seem like a second thought, and are mostly made of either reading long Jupyter notebooks, or running simple pre-coded ones to answer a short quizz. Statistical modeling is a topic that shoudl naturally lend itself really well to a "learn by doing" method, but unfortunately this course took the more traditional academic approach (nothing wrong with the later, it's just less engaging for me, especially when sitting in front of a computer).

I do not feel like this course had given me great knowledge, there is a lot of theory and almost none practice of python, specially in the last two weeks. Topics are interesting and they are good as an opener to learn statistics but there is not enough python about them. I am disappointed on this specialization (specially on this course), I only finished the course because it was the one left to complete the specialization.