返回到 Bayesian Statistics: Techniques and Models

4.8

星

313 個評分

•

92 條評論

This is the second of a two-course sequence introducing the fundamentals of Bayesian statistics. It builds on the course Bayesian Statistics: From Concept to Data Analysis, which introduces Bayesian methods through use of simple conjugate models. Real-world data often require more sophisticated models to reach realistic conclusions. This course aims to expand our “Bayesian toolbox” with more general models, and computational techniques to fit them. In particular, we will introduce Markov chain Monte Carlo (MCMC) methods, which allow sampling from posterior distributions that have no analytical solution. We will use the open-source, freely available software R (some experience is assumed, e.g., completing the previous course in R) and JAGS (no experience required). We will learn how to construct, fit, assess, and compare Bayesian statistical models to answer scientific questions involving continuous, binary, and count data. This course combines lecture videos, computer demonstrations, readings, exercises, and discussion boards to create an active learning experience. The lectures provide some of the basic mathematical development, explanations of the statistical modeling process, and a few basic modeling techniques commonly used by statisticians. Computer demonstrations provide concrete, practical walkthroughs. Completion of this course will give you access to a wide range of Bayesian analytical tools, customizable to your data....

Nov 01, 2017

This course is excellent! The material is very very interesting, the videos are of high quality and the quizzes and project really helps you getting it together. I really enjoyed it!!!

Jan 09, 2020

Excellent teacher and very well taught. Right amount of theory and programming combination. Made the subject easy to learn. Enjoyed it very much. Thank you very much.

篩選依據：

創建者 Jonathan

•Jan 01, 2019

Just finishing this class now......it is very good. Much better than the first one in this series. The videos and examples are better explained, and you leave with a solid understanding of Bayesian Analysis. When I signed up for this class I really wanted to know how I could use tools like MCMC to perform real analysis, and I feel like I got what I signed up for. Well done!

創建者 Hugo R C R

•Jun 19, 2018

Brilliant course! Very well organized and with useful study cases.Suggestion: It would be nice to have the same examples in Python using, e.g. Stan or PyMC.

創建者 Brian K

•Apr 01, 2019

Excellent course! This covered a large amount of material, but it was well organized, with a good number of problems to solve. Matthew Heiner does an excellent job with the lectures and explains things well. Coming from the frequentist worldview, I found this course to be a definite challenge, but well worth the time.

創建者 zhen w

•Jul 28, 2017

really like the content.

the R material in this actually changes my view towards R, so thanks.

創建者 Igor K

•Jun 13, 2017

This course is a perfect continuation of the Bayesian Statistics course by Prof. Herbert Lee. It's not only mathematically rigorous but also very applied. Excellent for the beginners to the Bayesian Statistics as it allows to start confidently using Bayesian models in practice.

Matthew Heiner is an excellent lecturer. Thank you.

創建者 Georgy M

•Apr 01, 2019

The second course of the great series. The knowledge and skills gained in this course allow to actually do statistical analysis on scientific data. The course is very clear, systematic and well presented. Thank you!

創建者 Benjamin O A

•Jul 08, 2018

This is a great course for an introduction to Bayesian Statistics class. Prior knowledge of the use of R can be very helpful. Thanks for such a wonderful course!!!

創建者 Seema K

•Nov 17, 2019

One of the best designed courses. The material and videos are very precise and informative. The quiz questions and assignment are very enjoyable. Thank you !

創建者 Arnaud D

•Dec 08, 2018

Really interesting course. The coding session are useful and can be use cases for lots of various situations.

創建者 Eugene B

•Jun 26, 2019

The course provided a lot of very helpful tools. However, I believe it was a bit too fast paced. Furthermore, there were certain topics which were not explained clearly -- for example, the discussion of the Metropolis-Hastings Algorithm and Gibbs Sampling was extremely confusing.

創建者 Yahia E G

•Jun 06, 2019

Really good intermediate introduction to bayesian analysis. I really liked how hands-on the course is. The last project was very useful as one will likely to face challenges and try to solve them especially if you use a rich dataset.

創建者 Chiu W K

•Jul 29, 2017

Informative but the pace is slow

創建者 Sandra M

•May 14, 2018

Good course, but the peer review process for the Capstone project in Week 5 is broken. Based on submissions to the course Forum in which multiple students have submitted their work on time but not received a grade due to lack of peer reviewers, this has been going on .

創建者 Sathish R

•May 21, 2018

This course is taught in a way that not useful for real world applications.

創建者 Jiasun

•Jul 20, 2019

Not enough depth.

創建者 Vladimir Y

•Nov 11, 2017

The course requires good understanding of Bayesian methods and linear modelling, something that is covered in previous course of this track from University of California Santa Cruz.

All quizes are quite easy to complete after watching the videos, but don't be fooled by this apparent simplicity - there is much more to the class than just that.

Capstone project is challenging and does put to test all of the topic discussed in class,

discussion forums are very helpful and also are extremely interesting to read.

I can strongly recommend this class to anyone who is interested in Bayesian Methods.

I've seen quite a few of similar classes on Coursera, but this one is the best, in my opinion, but also is the hardest one.

Do not miss out on Honors track, recommended supplementary reading and Capstone - those are the gems.

創建者 Oaní d S d C

•Jun 07, 2018

Excellent course. R usage straight from the beginning, a much useful addition to the previous course. It's very complete and when something mentioned and not explained further additional sources are recommended. Lot's of practical work and the final project I found amazing, a very practical approach that should prepare you to write reports and seriously analyse data. I would just recommend to put in the course prerequisites some basic R and some experience with statistics and probability. Although the course can be taken in isolation, the previous one is almost a prerequisite (if bayes thinking is new to you)

創建者 Cameron K

•Jun 07, 2017

An excellent introduction to the rjags package in R and using it to perform Bayesian analysis. The applied learning is supported by lessons in Bayesian theory, however, most of the learning is focussed on fitting, assessing and interpreting Bayesian models using rjags and the rjags language. The course is accessible if you have a passing familiarity with statistics and R. I have used traditional, frequentist statistical techniques for five years and I had no trouble completing this course without having done any Introduction to Bayesian Theory course - just jump right in!

創建者 Jens K

•Jun 14, 2020

This course is fantastic. The presenter is talking slowly and concise, and doesn't shy away from letting the simple things sink in before moving on. At the end it'll get complicated, sure, but this course takes the time every piece needs, where one thing builds on another. I'm glad I made it through the prerequisite course, which was rough. I've had some personal delays throughout this course, but it was easy to re-visit past lectures to refresh and move on. Very well didactically prepared and presented.

創建者 Jerry L

•Jul 05, 2017

This course fills an essential gap in learning Bayesian statistics, and provides concrete assistance in moving from theory to actual model writing in R and jags. Worth every penny, and then some more. However, the course requires a fairly high level of comfort with both general Bayesian statistics and the R language. I think it would benefit from a brief introductory lecture on jags syntax, as well as some additional worked problem examples.

創建者 Cooper O

•Aug 02, 2017

This course was fantastic. It combined detailed learning materials with frequent and comprehensive assessments. While managing to cover everything from the basics of MCMC through to the use of a number of different bayesian models. My only issue with the course was that the learning materials encouraged copy-pasting code and often didn't properly explain the choice of priors and other details about the chosen models.

創建者 Paul J

•May 28, 2020

Really a great course! It IS challenging, but the professor does a wonderful job. He also put a lot of thought into helping students learn. For example, when you get an answer wrong on a quiz, each wrong answer has an explanation WHY it was wrong to help you better understand your mistake. And each correct answer also has an explanation :).

創建者 Mr. J

•May 01, 2020

Superb.

This course with the MCMC Markov Chain Monte Carlo simulation filled in a critical piece of the statistic puzzle for me. Absolutely brilliant.

A key feature of excellence in the curse is the R code samples that directly parallel the course content. I hope it becomes the new paradigm for all code based instruction on Coursera.

創建者 Danilo I

•Jun 06, 2020

I would say the teachers are amazing. The subject was hard to learn as I'm not in the math and stat field, but I think the explanations were so well constructed that it allowed me to go ahead and even finish a statistical report all by myself. I hope this pair of bayesian teachers don't give up on us and keep doing this amazing job.

創建者 Sergio M

•Jun 06, 2018

Excelente curso. Da una introducción a los métodos de MCMC de una forma bastante sencilla y fe acompaña en problemas de regresión utilizando JAGS. Recomiendo este curso a todo aquel que tenga nociones de Estadística Bayesiana, pero que tenga pendiente los métodos avanzados para muestrear la posteriori de los parámetros.