An Intro to PyMC and the Language for Describing Statistical Models

<p>In our previous article on&nbsp;<a href="https://pub.towardsai.net/why-most-introductory-examples-of-bayesian-statistics-misrepresent-it-d2e12ac69278?gi=f468b252c728" rel="noopener ugc nofollow" target="_blank">why most examples of Bayesian inference misrepresent what it is</a>, we clarified a common misunderstanding among beginners of Bayesian Statistics. That is, the field of Bayesian Statistics IS NOT defined by its use of Bayes&rsquo; Theorem but rather by its use of probability distributions to characterize uncertainty and consider the full range of possible outcomes. So, for example, rather than being told that a given medical device is 95% effective in detecting a disease&nbsp;<em>given</em>&nbsp;the fact that you know with absolute certainty you are already infected (what we can otherwise call our&nbsp;<strong>True Positive Rate (TPR)</strong>), we can consider scenarios where that device is only 24%, 69%, or 91% effective in detecting the disease and how those numbers change your probability of contracting it&nbsp;<em>given&nbsp;</em>that you have tested positive from the device.&nbsp;</p> <p><a href="https://pub.towardsai.net/an-intro-to-pymc-and-the-language-for-describing-statistical-models-ac9d56c363c0"><strong>Click Here</strong></a></p>