While My MCMC Gently Samples

Bayesian modeling, Data Science, and Python

An Intuitive Guide to Bayesian Statistics

Are you a software engineer wanting to move into data science or a data scientist looking to stand out from the crowd?

Have you found most statistics books overly theoretical, math-heavy and without a clear focus on application and intuitive understanding?

Do you want to use these crazy times to keep your skills sharp to improve your career prospects?

Then this course is for you.

Let me start with a confession: I've never been great at math. Learning statistics took many frustrating hours to understand the math that was thrown at me. Once I did intuitively understand the concepts they suddenly felt trivial and I thought "well why didn't they just say that!".

If, like me, you have a coding background, math will not be your first language. You think in code and figure things out by playing around with them.

I have suffered through enough dry math so that you don't have to, and that is what I have done with my blog and talks I've given. For example, my post on Markov chain Monte Carlo sampling gets by with 3 simple formulas but plenty of code and visualizations to explain the intuition of the algorithm, which is much simpler than the math lets on. See the quotes at the end of this post for the feedback I've gotten on this teaching style.

Now I'm not saying that math is not important or won't be covered in the course, but it will always come second, understanding and intuition is much more important. And once you have an intuitive understanding, the math suddenly becomes trivial.

This course will consist of short videos explaining key concepts of Bayesian modeling with a heavy focus on application. See this post for why Bayesian statistics is such a powerful data science tool. We will make use of Probabilistic Programming tools like PyMC3 which allow easy specification of statistical models in computer code without deep knowledge of the underlying math.

If you want to keep updated about the course, get a sneek-peak, and lock in an early-bird discount, enter your email here:

Quotes

"If all algorithms were explained as clearly as this, I'd be so much further ahead in having all of these statistical tools in my mental toolbox." Robert M Johnson

"Being completely new to the Bayes’ theorem and MCMC algorithm, I have found the article illustrative, educative, and well understandable! Great job Thomas!" Vaclav

"Excellent post, this is the most intuitive explanation about MCMC I have ever read. I really like how you use code instead of math to explain the algorithm." Jun

"Excellent writing. Finally I understood the MCMC. Thanks a lot."

"Great writeup Thomas. I'm a huge fan of your focus on the intuition behind the math." Brandon Rohrer