Skip to content

MwM: *Why* don't I get Bayesian statistics? #477

Description

@craig-parylo

Bayesian statistics is often presented as a powerful alternative to traditional statistical methods, but for many analysts trained in a frequentist framework, it can feel confusing, unintuitive and difficult to approach.

In this session, I'll share my own attempt to understand Bayesian statistics from the perspective of a healthcare analyst who has spent their career working with more traditional statistical methods. Rather than presenting a comprehensive technical introduction, this talk explores the barriers I encountered and the insights that helped me start making sense of Bayesian thinking.

We'll look at some of the common stumbling blocks, including:

  • Navigating unfamiliar terminology such as priors and posteriors
  • Understanding conditional probabilities and why they can be surprisingly difficult to reason about
  • Overcoming the intimidation of Bayesian formulae and mathematical notation
  • Exploring why Bayesian results often seem to contradict our intuition

No prior knowledge of Bayesian statistics is required. The only prerequisite is a willingness to have your intuition challenged.

Metadata

Metadata

Assignees

Type

No type

Projects

Status
Scheduled

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions