If you have a pos. test in 9000 cases people actually have disease BUT in ca. 50000 cases they don't have the disease -> low probability of having the disease when you have a pos. test

A test is sensitive when it can with high probability correctly diagnose having the disease

Invert the generative model (is key for understanding the base rate neglect) -> Given the observed data, what can we conclude about the state of the world

Given a particular state of the world, these are our predictions

-> p-values...

stacked log. regression models

A deterministic component refers to the predictable, non-random part of a system or signal. It represents the part that can be calculated or predicted with certainty, given the initial conditions and rules of the system. This contrasts with stochastic (random) components, which introduce unpredictability.

refers to data or observations that are unreliable or unclear, making it difficult to draw definitive conclusions -> any information that is difficult to interpret due to its lack of clarity or reliability