Good intro to Bayesian Statistics. Covers the basic concepts. Workload is reasonable and quizzes/exercises are helpful. Could include more exercises and additional backgroung/future reading materials.
An excellent course with some good hands on exercises in both R and excel. Not for the faint of heart mathematically speaking, assumes a competent understanding of statistics and probability going in
par Joseph R R•
Liked the course, but it was a little easy (took four days total to do the material for the whole course). Many questions were left unanswered (such as how dependent the credibility intervals are on the choice of prior distribution and the assumed distribution of the data), and it didn't touch on later topics that are interesting (MCMC sampling). Again, good beginning course, but I was looking for more in depth study.
par Tianchi L•
-1 star: Some discussions and derivations do not have adequate context and background. I expected more thorough explanation on concepts and more advanced topics. There are also a few minor typos that confused me. It is only a helpful introductory level course on Bayesian without depth.
-1 star: quizzes are not challenging enough and they only require plugging in numbers into equation. Not a good way to study
par A l•
The first two weeks are very clear, after that, new notions are thrown without any definition, the calculations are not done, only results are given. I finished the course by brute-forcing the exams because I wanted to finish fast to take another course... No help in the forums too. For me this is a course to avoid except the first two weeks that helped me a lot.
par Francois S•
Nice introduction to Bayesian concepts. Presentation sometimes focused on the details of the calculations and could gain from more perspective. Sections relating to Normal variables - variance unknown and Linear Regression could be more explicit. Useful overall as an introduction, but require to get additional external material to get to the bottom of it.
par Jens R•
It was pretty intuitive and easy to follow the first couple of weeks, but then the assumed knowledge of beta and gamma distributions and their frequentist usage, stood in the way of me fully grasping the Bayesian part of it. In the end I just copied the examples from the lectures and passed the tests ... without really getting it.
par Edoardo C•
Overall I liked the course but I would have preferred a more formal treatment in many cases - sometimes numbers were plugged into the formulas without first explaining their formal structure more in detail.
I did not like also the fact that the course was implemented in R and Excel (but that's a matter of taste of course).
I didn't think the lectures were very good. The instructor wasn't careful with his notation, which was very confusing, and the initial lectures where he used a pastel green marker on a green background and wearing a pastel green shirt made his blackboard text nearly invisible.
However, the assignments were execellent.
par Dmitry S•
The material is good, but I've found the lectures challenging to understand even having some background in math. It would be good if all the definitions and key facts were stated more prominently in the lectures, as opposed to algebraic transformations which most readers can hopefully do on their own.
par Ahmed S•
This course requires solid grounding in mathematics. No meant of Social Science graduates without proper training in statistics/mathematics. The course was good in the sense that we could how probability distributions are used to model real world problems.
Study material was certainly not adequate.
par Ray P•
The course presented fundamental concepts of probability, regression, and Bayesian ways of thinking. However, it lacked in applications of Bayesian approaches beyond the most basic. For example, how do we estimate these models on real data to obtain parameters and make inferences or predictions?
par Yuzhong W•
The lectures from week 1 to week 3 are nice and useful to me, but I think there should be more details about the content in week 4. For example, I think the lecture about the Jeffreys prior skipped many things and I did not understand this concept very well.
par Damel L•
Most of the support material should be prior reading. Lecturing could be more useful i.e. explaining ore about why we use certain distribution and how to apply them. Most of it as just reciting formulas and felt like a waste of time...
par Olexandr L•
It was quite difficult to learn from just the material provided here, and I had to look for info on the web. Also, adding modern real life examples and going into detail would make this course better
par Jesús R S•
Good course as an introduction to bayesian statistics if you want to pursue more advanced courses in the field or to get some practise working with distributions under the bayesian framework.
par Silvia Z•
In general, the course is useful, but in half of videos the explanation focused mostly on formulas, and less on theory. I personally had difficulty in learning theory of Bayesian statistics.
par Borja R S•
The teachers are clearly experts in what they do, but sometimes I think it is that same expertise that makes them jump to conclusions too easily, making it difficult for beginners to follow.
par Ran W•
This course gives a very brief background on conjugate prior. However, the lectures on Bayesian linear regression is too superficial. I wish the lectures could have gone into more detail.
Too much time spent on the beginning and too little on later more complicated concepts such as the posterior predictive. It felt as if that was just a side note in the extra readings.
par Augusto S P•
The course is good for beginners in statistics. In my opinion it would be better to invest more time explaining different topics about bayesian regression and bayesian time series.
par Oliver B•
Solid mathematical grounding, but would have benefited from more time spent on the history of Bayesian inference, when to use it, why it can be used etc..
par Pranav H•
The course could have given more information on tiny details which can confuse people during the exercises. But overall a good learning experience
par Ángel L•
It’s ok to have a theoretical basis about Bayesian Statistics, but I missed some practical cases using Python instead of R. I also missed PYMC3
par Kathryn L•
It's a nice introduction to the topic, but I often found the lectures to be imprecise or inconsistent, especially with respect to terminology.
par Alessandra T•
We still don't understand how Bayes differs to Frequentist... A worked example comparing the two at the end would have been nice.
par Ken M•
It would have been great if more graphs had been provided, for easier visualization of the e.g. distributions, or concepts.