Category: Uncategorized
-
Ambition, Distraction, Uglification & Derision + MRE
I’m looking forward to reading Alice in Wonderland to my daughter some day. I honestly don’t remember a lot of the book, but there is a scene at the beach where the Mock Turtle is teaching key subjects: “Reeling and Writhing of course, to begin with,’ the Mock Turtle replied, ‘and the different branches of…
-
Golems: Statistical Rethinking Wk 1
The first lecture of 20 in Richard McElreath’s course, Statistical Rethinking, is introduces the metaphor of a clay Golem from Prague which is eventually destroyed by its creator — for statistical modeling, McElreath’s point is that whatever you create, no matter how powerful, it cannot think for you and so you had better hope to…
-
Waterfall Charts are great pricing visuals
The floating bars help the viewer to understand that each element is incremental and the ends represent the running sum. Did you know its super easy to make waterfall charts in MS Excel now? These used to require a lots of formatting and careful structuring of your data, but now its a predefined chart type…
-
Color Ranges: Another use for percentile rankings!
Sometimes I see how other people do things and I think, “that’s so obvious, why didn’t I think of that?”. Yesterday I had one of those moments when I saw that instead of re-centering the color range you can ensure your Viz uses the entire color palette by creating a measure for percentile rank of…
-
Moneyball – Feature Engineering
I first read this book in college because our catcher recommended it to me. It was the first Michel Lewis book I ever read and I happened to be in my first year of studying economics so it set off all sorts of mental connections. I was thinking about Slugging Percentage, which I first heard…
-
Magical Thinking: Data Agnostic Algorithms
Today I had a good chuckle with a colleague about why the Data Science team couldn’t use a Dev database with a limited subset of sample data in the schema of Prod to develop their machine learning models — the amount and complexity of the data you have completely drives the predictive ability of your…
-
“All models are wrong but some models are useful” George Box
I started re-reading the Alignment Problem and saw that the author Brian Christian references Box too. He gives the cautionary tail of the sorcerer’s apprentice, bewitching brooms to haul water on his behalf, carrying out their prescribed task to a T with an undesirable result, to illustrate the problem of ill-defined objective functions. In my…
-
Automation vs Manual Labor Trade-off
Sometimes the most valuable way to contribute to an automation project is entirely manual. Two pricing projects I did entirely manually that potentially could have been done with machine learning techniques that had tremendous value, but required very little skill. 1)Culling through paper contracts and PDFs to find any redlines or deviations in our standard…
-
What is easier to teach? Domain knowledge or Data Science?
Does it make more sense to teach statistical methods to domain experts or domain knowledge to your data scientist? For all the growth in the data science field I don’t think the problem solving approaches from dedicated data science and statistics programs have been fully integrated into the training programs for other fields of study.…
-
Admiration & Gratitude
Simpson’s Paradox, Anscombe’s quartet and the Datasaurus Dozen are all posts about statistical summary concepts with corresponding Tableau Public workbooks created by my colleague Alexander Mou on his blog “Vizible Difference”. Working with people who are intellectually curious, self-motivated and exceptionally skilled in their trade is a source of joy for me. I love that I…
Carragee Consulting 