Musings, Explorations, and Announcements

Here are my notes from a recent talk I gave on vectorization at a Davis R Users’ Group meeting. Thanks to Vince Buffalo, John Myles White, and Hadley Wickham for their input as I was preparing this. Feedback welcome! Beginning R users are often told to “vectorize” their code. Here, I try to explain why vectorization can be advantageous in R by showing how R works under the hood. Now, remember, premature optimization is the root of all evil (Knuth).

Read More…

I just submitted the abstract below to the ESA 2014 meeting. Can’t wait! I’m also organizing an oral session with Richard Cobb: “Ecosystem and Community Effects of Native and Invasive Diseases.” Click through for more details. It’s gonna be a good year! I haven’t been to ESA since 2008. Modeling forest disease using a macroparasite framework Noam Ross, UC Davis Graduate Group in Ecology Background/Question/Methods: Forest pathogens are typically modeled using variants of SI models, in which the infection states of trees are binary.

Read More…

Today at the Davis R Users’ Group, Bonnie Dixon gave a tutorial on the various ways to handle dates and times in R. Bonnie provided this great script which walks through essential classes, functions, and packages. Here it is piped through knitr::spin. The original R script can be found as a gist here. Date/time classes Three date/time classes are built-in in R, Date, POSIXct, and POSIXlt. Date This is the class to use if you have only dates, but no times, in your data.

Read More…

Yesterday at the Davis R User’s Group¹, Ryan Peek gave a talk about using the shiny package to create interactive web apps with R. Here are his slides. Ryan includes a bunch of links to examples and tutorials, as well as his own thermohydrographs app:

---

1. Thanks to Revolution Analytics for another year of sponsorship!↩

Many ecological processes are best represented by individual-based models (IBMs), where processes at the scale of individual scale of organisms and their actions are modeled explicitly. Such processes can pose a problem for management or control; our observational scale, control levers, or management goals may be limited to coarser scales than the individual. IBMs suffer from the curse of dimensionality. The number of possible states of the system, which consists of the possible states and combinations thereof of all individuals, is enormous.

Read More…