How to Learn Probability Distributions

The machine learning consultancy: truetheta.io/
Join my email list to get educational and useful articles (and nothing else!): mailchi.mp/truetheta/true-theta-email-list
Want to work together? See here: truetheta.io/about/#want-to-work-together

In this video, I share a perspective on probability distributions that makes understanding and retaining them easier.

SOCIAL MEDIA

LinkedIn : www.linkedin.com/in/dj-rich-90b91753/
Twitter : twitter.com/DuaneJRich

Enjoy learning this way? Want me to make more videos? Consider supporting me on Patreon: www.patreon.com/MutualInformation

Sources and Learning More

The incredible chart mapping out the relationships among probability distributions comes from Larry Leemis and Jacquelyn McQueston[1]. They and many other contributors put together a website providing all the details on that chart : www.math.wm.edu/~leemis/chart/UDR/UDR.html

Wikipedia has two [2][3] articles that provided a lot of interesting relationships. Also, John D. Cook proves that the student-T is a mixture of normals in [4].

[1] Leemis, L. & McQueston, J.T. (2008), Univariate Distribution Relationships, The American Statistician, Vol. 62, No. 1

[2] Relationships among probability Distribution, Wikipedia, en.wikipedia.org/wiki/Relationships_among_probabil…

[3] Compound Probability Distribution, Wikipedia, en.wikipedia.org/wiki/Compound_probability_distrib…

[4] Cook, J.D. (2008), Student-T as a Mixture of Normals, www.johndcook.com/

Extra Notes

In this video, I label the Poisson a "continuous" distribution, which is certainly not true in the general context - it is as discrete as they come. A better label than "continuous" would directly reference the specific limit that is taking place. Also, such a label would separate them from the other continuous distributions that can be reached with different limits. Thank you mCoding for pointing this out.