Publications by John Mount
Changing Forecasts for Python Questions on Stack Overflow
I recently conducted a small time series workshop session for AI+ training hosted by ODSC. It went really well, and I’d be happy to offer longer interactive workshops going forward (please reach out if your team would like one!). One of the examples I shared was derived from the following graph presented in The Incredible Growth of Python on Stac...
2849 sym 6 img
Best Before Dates by Bass
I was searching for one last real world example for my upcoming video talk March 13th on time series forecasting. Hope to see you there! Or reach out to Win Vector LLC for custom training! I had the seemingly harmless thought: “Let’s look at Stack Overflow trends“. In particular their pre-built “data science and big data trends” query s...
2446 sym 8 img
Kelly Can’t Fail
Introduction You may have heard of the Kelly bet allocation strategy. It is a system for correctly exploiting information or bias in a gambling situation. It is also known as a maximally aggressive or high variance strategy, in that betting more than the Kelly selection can be quite ruinous. I recently ran into a strange card game where the Kelly s...
6738 sym Python (1495 sym/6 pcs)
Reposting Partial Pooling
Nina Zumel had some good articles on partial pooling estimators that I want to return to. It is a great technique to get more reliable models when using categorical variables. I wrote an introduction to them here some time ago. More importantly Nina has now repaired the damage to the mathematics rendering we picked up in a WordPress migration. So p...
913 sym
Examining Meta-Analysis
Joseph Rickert and I put together an experiment trying to both run a standard meta-analysis and then reproduce similar results directly using Bayesian methods. I think it came out really interesting and we share it here at R Works and also here on Github. A meta-analysis is an attempt to estimate an effect by combining several pre-existing studies....
1680 sym 2 img
Calculating at Pencil and Paper Scale
Introduction It can be fun to drive a problem all the way into the ground. I don’t always get to do that on paying projects, however sometimes I can do it with hobby projects. In this case I am going to re-solve Dudeney’s Remainder Problem again and again to argue that it can be solved at a pencil and paper scale. The problem is as follows. A ...
6439 sym 8 img
Dyson’s Algorithm for the Twelve Coins Problem
Nina continues with the 12 coins problem by transcribing Dyson’s algorithm into R. It is kind of a fun article. Most of us see the 12 coins problem as a one-off puzzle that we spend a little time with and give up on. In her earlier “The Twelve Coins Puzzle” Nina used an algebraic notation to allow her to search long enough to solve the 12 coi...
2026 sym
Come Learn About Time Series with Me at ODSC West Wednesday October 30th, 2024
I’ll be teaching how to solve time series problems at ODSC West Wednesday October 30th, 2024. This is a neglected topic that can unlock a lot of value in your organization.Find at https://odsc.com/california/odsc-west-schedule-2024/ , Tutorials | Workshops, 30 Oct Day 2. 3:30 PM Pacific time, 1 hour long. Hope to see you there!!!(Hyatt Regency Sa...
597 sym 1 img
Dudeney’s Remainder Problem
The remainder problem The description of this puzzle really cracks me up (Dudeney, Strand Magazine, January 1924). Health risks aside, how do we find the maximal integer d such that (480608 % d) = (508811 % d) = (723217 % d)? The solution A good puzzle strategy is to try to simplify the problem. This particular puzzle is equivalent to solving: 50...
1695 sym R (298 sym/1 pcs) 4 img
An Easy Puzzle: The Perplexed Banker
Nina Zumel continues with the puzzles. This one is “The Perplexed Banker”. In my opinion, this one captures the essence of the “mathematical” aspect of a puzzle. For a mathematical puzzle one often hopes there is a systematic method that makes the puzzle easy. In this case there is indeed a systematic method that makes the puzzle easy. Then...
868 sym