Bayesian Statistics: A Step-by-step Introduction

What you’ll learn
Understand how Bayes’ rule can be used to update beliefs
Use conjugate priors and likelihoods to model binary, count, and continuous data
Understand the concepts of prior distributions, posterior distributions, likelihood functions, and predictive distributions
Understand how statistical software can be used to compute and visualize information about your beliefs

Requirements
Strong skills in basic algebra and arithmetic
Some knowledge of calculus is useful, but not required.

Description
This course teaches the foundational material of statistics covered in an introductory college course, with a focus on mastering the basic components of any Bayesian model – the prior distribution and the likelihood, and how to find a posterior distribution, credible intervals, and predictive distributions.  Along the way, you’ll become more comfortable with probability in general and gain a new perspective on how to analyze data!We start from scratch – no experience in Bayesian statistics is required.  Students should have a strong grasp of basic algebra and arithmetic.  R and RStudio, or Python, is required if you would like to run the optional coding sectionsThe course includes:5.5 hours of video lecturesInteractive demonstrations using R and Stan (Python code is included too!)Quizzes to check your understandingReview assignments with solutions to practice what you have learnedYou will learn:The basic rules of probabilityBayes’ rule, including common examples with medical testing and flipping coinsThe terminology of different components of a Bayesian model: the prior distribution, posterior, likelihood, and predictive distributionConjugate priorsCredible intervals and Bayes estimatorsModeling binary data with the Bernoulli and Binomial Distribution, and the Beta distribution priorModeling count data with the Poisson Distribution, and the Gamma distribution priorModeling continuous data with the Normal Distribution, and the Normal distribution priorAn introduction to simple linear regressionThis course is ideal for many types of students:Anyone who wants to learn the foundations of Bayesian statistics and understand concepts like priors, posteriors and credible intervalsData science and data analytics professionals who would like to refresh and expand their statistics knowledgeAcademics in the social, biological, and physical sciencesThis course is ideal for anyone, from beginners to seasoned professionals. It doesn’t matter if you’re just starting your journey in data science, looking to upgrade your existing skills, or simply have an interest in Bayesian statistics. My goal is to make Bayesian statistics accessible and understandable for all.

Who this course is for
Current and aspiring data scientists and data analysts
Academics in the social, biological, and physical sciences
University students studying mathematics or statistics
Anybody who wants to learn to rigorously update their beliefs from data.