Computational Methods for Numerical Analysis is Out Now!

My new book Computational Methods for Numerical Analysis with R came out today from Chapman & Hall/CRC Press. This book is a long-standing project of mine–originally started in 2001, and then using Octave as the base language. As the world didn’t need yet another book on doing numerical analysis in Octave, and therefore MATLAB, I eventually moved it to R. Computational Methods for Numerical Analysis with R is an overview of traditional numerical analysis topics presented using R. This guide shows how common functions from linear algebra, interpolation, numerical integration, optimization, and differential equations can be implemented in pure R

Implications of Coarse Data Allocation Methods for Flood Mitigation Analysis

I went looking for a copy of my working paper from the 2013 Joint Statistical Meetings and realized, it does not seem to be online. And neither does my poster! So I figured I should fix that. Here’s a link to the poster and the proceedings paper. The key idea here comes from my work on flood insurance and trying to understand the county-level effects of flood mitigation grants. Some grants are given to state agencies. In other words, the granularity of reality is insufficient for my goals. Normally, it’s just a data problem. So I investigated, briefly, four logical

Data Science for Restaurant Inspections

At work, we just did a really neat set of predictive models for restaurant inspections. This is all based on the work Chicago did for the analysis. We kinda/sorta split into different groups and did analyses for three cities (with links to reports): Raleigh, North Carolina, Syracuse, New York, and Denver, Colorado. Together, these three reports show different approaches and analyses we used in the three different cities, along with discussion of how we applied the Chicago work. More information is available from our GitHub page: GitHub – iscoe/restaurant_inspections: Predicting violations for restaurant inspections restaurant_inspections – Predicting violations for restaurant

Announcing Computational Methods for Numerical Analysis

For about two years, I’ve been working on a book called Computational Methods for Numerical Analysis with R (CMNA), which will present an outline of numerical analysis topics with original (and simplified) implementations in R at a level appropriate for a graduate student or advanced undergraduate. Last night, I sent the latest draft to my editors, and I am quite pleased to say it should be heading into production, soon. The organizational structure of the text is based roughly on the organizational structure of MAPL 460 15-20 years ago: Introduction to Numerical Analysis Error Analysis Linear Equations Interpolation and Extrapolation

Deep Analytics and Big Data

Bill Marcus writes about the intersection between deep analytics and big data in HPE Insights: [T]he scale of an organization and its data is key to how much the process will matter, according to James Howard… “If you’re a small bank, machine learning is going to help get you through the day, but it’s not going to be something critical—whereas if you’re doing high-frequency trading, all you’re doing is machine learning.” Read more at HPE Insights: In the minds of machines: Fundamental change from deep analytics – HPE Business Insights By Bill Marcus, contributing writer In Munich, Germany, the technology

Of Course NaN^0 = 1

David Smith and I are now talking to each other in blog posts and it is only a little weird. Also, I’ve been traveling and am a bit behind. In a comment on this post, he notes this: I suspect the reason why R Core adopted the 0^0=1 definition is because of the binomial justification, R being a stats package after all. I can’t think of any defense for NaN^0=1 though… Well, it turns out there’s a good reason. If we go back C, and try an experiment, we can observe the following example produces these results: Compiling and executing