There were a few instances where wrong numbers in equations/slides (if everything is not correct, obviously I get confused as the learner), and there are some questions on the quiz that have not been addressed properly in the slides (i.e. using tables to look up z score values). Also lectures are incomplete (especially no lectures on MAP Estimation when previous videos clearly talk about it and it's even on the quiz). Would not recommend this course to anyone until it's been fully built out/properly reviewed.

While the outline of the course is promising, the explanation is far from reasonable or clear. Few things in the course that drove me nuts:

- Using several terminologies in the same exact video to refer to the same exact thing. For example: sigma + standard deviation, Gaussian distribution + Normal distribution. While technically this is not wrong, it is just confusing to keep up with what the instructor is referring to while you are trying to learn new concepts at the same time

- Using non-standard notation. In many videos the reference for mu, E[X] and variance used some notation that I did not manage to get hold on or find them being widely used in any other materials.

- Incorrect notations, beside the none standard, there were instances (I reported them to be reviewed) where the notation is plain wrong. For example summation over n = 1 to n = N of (E[X] - mu_x)^2 * P(x_i) What does it mean to have subscript for mu? do we have more than one mean? also where does the subscript i came from?

I am in the second week and I am sure I won't pass this course unless I used other learning materials to get better handle on the concepts

This course gave me the impression that it was not created with the same attention as the other two in the specialization. There is a lot of things to learn but the lessons felt too fast and compressed.

Thdis course should either reduce the arguments or take its time to explain with longer/more videos and more intermediate labs and tests to make the experience more engaging.

A good primer, but not the best to explain concepts. Putting a course together is one thing and actually making someone understand the concepts is a completely different animal. Very difficult to excel in making concepts clear.

This was another excellent course in an absolutely fantastic specialization track. The amount I learned per hour spent is far above any other specialization I've ever done. This will help me immensely in terms of career preparation.

Despite Luis's significant knowledge and teaching experience, the course wasn't as clear or easy to understand as it could have been. There were often chances for him to provide high level intuition behind new concepts like p-values or Z-statistics that he neglected to do, or did in an unintuitive way. And there were a number of verbal typos throughout the videos. A few specific points of feedback: Week 1 Lesson 2 - probability distributions Explanation of Bernoulli distribution was poor * Should have explained that a Bernoulli distribution is a special case of the binomial distribution with n = 1 Week 3 Lesson 1 - Central limit theorem - straightforward concept but really convoluted explanation. I found chatGPT’s explanation and example much more straightforward. Week 4 Lesson 1 - video 1 “For illustration's sake, imagine that you were taking a sample of size 1 and finding the mean of that sample to use as your estimate for the population mean. From the central limit theorem, you know that if you were to take multiple samples of this size and create a sampling distribution for the sample means for a sample of size 1. The sampling distribution of the sample means makes a gaussian with a center at mu and a standard deviation of sigma.” —> This is not true unless the underlying distribution is already normally distributed. Week 4 Lesson 2 - p-value lesson felt really rushed, especially the part on Z-statistic (unclear how to actually use it)

Excelent course for those who want to get a deeper understanding on statistics and advanced analytics

Probably the most difficult course in this specialization. Statistics and Probability is the branch of mathematics that challenges your intuition about data. But completing this felt rewarding. Well explained. Great labs.

The instructor mistakes a lot. Lots of information weren't explained clearly and sometimes skipped

It was a super exciting journey through maths. My last courses in my were 20 years ago, and it was easy to follow and remember all these topics.

The programming assignments were challenging. Especially Week 1. Worth the effort.

Overall a solid course but there are things which could be made better. Let's talk about those first. I think the first half of the course is much better than the second. There are more in-video quizzes, and concepts are unpacked more intuitively. In the second half of the course a lot of times videos felt rushed, lots of formulas are just dumped on the screen without really explaining them and there is too much alternative (unexplained notation going on). In addition, some of the more complex topics just have reading, no video (like MAP). On the positive side most of the time Luis does a good job at explaining stuff intuitively and explaining how it relates to machine learning. Concepts are covered from multiple angles and are built from the ground up. The assignments are quite interesting and include exploring the birthday problem and A/B testing. Topics covered are quite broad: probability, Bayes theorem, hypothesis testing, confidence intervals, samples and population and probability distributions.

Need more exercises to practice

Extraordinary course. With clear explanations and animation video. I learned Probability and statistics before but forgot a lot. This course helps me reinforce my knowledge about this subject as well.

Excellent course. I studied probability and Stats long ago in university, but this course covered it in far greater depth.

It was a perfect course

Great materials, but would like more real-world examples

There are some confused terminologies to me

This course has a great quality materials when it comes to graphics and animations. In some places, it also presents the concepts in a really intuitive way (like showing that regularisation can be thought of as a maximising conditional probability). But in my opinion, while creating the course, too much focus was put on the graphical part of the course, which is slightly overdone, while the materials are confusing in some parts (like in the Law of Large Numbers and the CLT, where the assumptions are shown in a way that's just obfuscating the concepts). Also, the course is way, way much overdone in terms of being beginner friendly. For example, the instructor spent tons of time on explaining trivial concepts, like expected value, illustrating it with multiple examples and lots of animations, but went really quick through more interesting ones, like the Central Limit Theorem. I personally found the course boring in the parts when the simplest concepts were presented and confusing in the parts describing more difficult ones.

Great course but the stats aspect was not so clear, also labs should be in the videos and instead of do on big lab assignment at the end, it should be in intervals, so after 4 videos do an assignment. But over all good course

Material was a bit rushed

Let's start with the positives: Some of the ungraded lab are great introductions to data analysis and application of simple math to machine learning models. They are easy to understand, fun to play with and you can see some time has been put into making them useful. That's unfortunately about it. While I was incredibly happy with the first two sections of the specialization and used it to implement my college studies on the matter, this part feels incredibly rushed. Lots of formulas are thrown in and not explained. Luis, a great teacher in the first two courses, seems exhausted and just trying to spit out a list of facts. Most things are thrown at the student without a clear path or link to the examples shown. The graded labs are either too easy or leave the student with no help. This results in a lower quality specialization on the hardest of the three topics. Not much is retained after the course unless you are studying the subject on your own. On top of this, the quality of the videos is very low compared to the first two parts. Continuous disturbing sounds, incorrect formulas, dropped in voiceovers make it frustrating to follow. I really hope they can improve it and make it as good as the first two parts.

First week was ok at explaining the concept. I found w2-w4 difficult to understand and I ended up watching other YouTube videos (StatQuest, Brandon Foltz, 3Blue1Brown) which did a much better job at explaining the concept and material in a much simpler and intuitive way. This course needs some improvement. The instructor speaks too fast in the video. It's difficult to follow along when you try to read the notation. The labs do help to understand the concept in a practical way.

Unfortunately, I cannot recommend this course. It won't suit beginners because the instructor lacks the skills or desire to explain concepts effectively. Similarly, it won't suit advanced learners due to the shallow material. Simply reading formulas from the screen does not constitute effective teaching. However, the labs are enjoyable.

Sin ninguna duda, el peor curso que he tomado. Esta especialización de 3 meses fue una perdida de tiempo. Me imagino ahora que hubiera pasado si esos tres meses hubiese tomado un curso que de verdad valga la pena. Me la pase más viendo videos de youtube que explicaban mejor los temas y ChatGPT que daba mejores ejemplos. Sin ninguna duda no recomiendo este curso, 1 estrella es mucho para este curso.