Data Science Bite: What is Statistics?
Statistics is the science of relating summaries of observable samples to the unobserved summaries of the populations they are drawn from. I try to explain that with an example in this video. (link)
Data Science Bite: A/B Testing
I’d like to share my latest “data science bite“: A/B Testing.
Upcoming Talk on Probability Models
Nina Zumel and John Mount will be speaking at the online University of San Francisco Seminar Series in Data Science! How and why to use probability models to outperform decision rules Friday April 30, 2021 12:30pm – 2pm Pacific Time See here for full details and to RSVP In this […]
Data Science Bite: “What is Supervised Machine Learning?”
I am trying a new idea: “data science bites.” Data science bites are small articles and videos explaining only one idea each. This first one explains what supervised machine learning is, without going into the details of how it is realized. (link)
When Profitable Betting Systems are not Possible
I felt a bit guilty explaining a Kelly/Thorp style card betting system without discussing why these ideas don’t work on fair coin games. So I have “writeup for engineers” on the martingale theory of such games. This has example code, so one could try to come up with a betting […]
Kelly Thorp Betting
I demonstrate a Kelly/Thorp betting system for the simple card game of guessing if the next card from a standard deck is red or black. I have a video of the play here. And a derivation of the betting strategy in R is here. A derivation of the proof you […]
Plotting Multiple Curves in Python
I have up what I think is a really neat tutorial on how to plot multiple curves on a graph in Python, using seaborn and data_algebra. It is great way to show some data shaping theory convenience functions we have developed. Please check it out.
It Has Always Been Wrong to Call order on a data.frame
In R it has always been incorrect to call order() on a data.frame. Such a call doesn’t return a sort-order of the rows, and previously did not return an error. For example. d <- data.frame( x = c(2, 2, 3, 3, 1, 1), y = 6:1) knitr::kable(d) x y 2 […]
Introducing wrapr::bc()
The wrapr R package supplies a number of substantial programming tools, including the S3/S4 compatible dot-pipe, unpack/pack object tools, and many more. It also supplies a number of formatting and parsing convenience tools: qc() (“quoting concatenate”): quotes strings, giving value-oriented interfaces much of the incidental convenience of non-standard evaluation (NSE) […]
What is a Good Test Set Size?
Introduction Teaching basic data science, machine learning, and statistics is great due to the questions. Students ask brilliant questions, as they see what holes are present in your presentation and scaffolding. The students are not yet conditioned to ask only what you feel is easy to answer or present. They […]