Trevor Hastie: Statistical Models for Presence-Only Data (Friday 3rd May 2-3)

Trevor Hastie is a world-renowned statistician with major contributions to applied statistics, including two classic textbooks. His current research focuses on applied problems in biology and genomics, medicine and industry.

We have selected a paper co-authored with William Fithian that is sure to be a citation giant for ecologists and statisticians alike, addressing many key challenges in analyzing presence-only data: "Statistical models for presence-only data: Finite-sample equivalence and addressing observer bias".

This is a heavy-hitter, so we will focus mostly on Sections 1, 2, and 5.

Feel free to post any questions and/or comments.

Code for spatstat Lab

Here is some basic code to get started with spatstat using the Beilschmiedia pendula dataset.

install.packages("spatstat")
library(spatstat)
data(bei)

plot(bei)
plot(bei.extra$elev)
plot(bei.extra$grad)

fit.1 = ppm(bei, ~poly(elev, grad, degree = 2), covariates = bei.extra)
plot(predict(fit.1))
plot(bei, add = TRUE, cex = 0.4)
diagnose.ppm(fit.1)
AIC(fit.1)

fit.2 = ppm(bei, ~poly(elev, grad, degree = 2) + poly(x, y, degree = 2), covariates = bei.extra)
plot(predict(fit.2))
plot(bei, add = TRUE, cex = 0.4)
diagnose.ppm(fit.2)
AIC(fit.2)

Can you find a model which removes the main trend in residual plots?

Adrian Baddeley: spatstat Lab Friday Apr. 26

This Friday we aim to understand a bit more about the work of our next speaker, Adrian Baddeley. Adrian is an expert in spatial point patterns, and his freely-available R package spatstat has taken the world by storm!

To get a sense for what spatstat can do and how to use it, we will be taking a look at the lecture notes available at the following link: http://www.csiro.au/resources/pf16h. In particular, Chapters 1 and 2 offer a nice overview of what spatial point patterns are and what types of questions can be answered using spatstat. Taking a look at any other Chapters of the lecture notes is encouraged, but not essential for what we’ll discuss.

At UNSW,  the plan this week is a little different – we’ve booked a lab in Bioscience Room 640 for Friday from 2-3 to get familiar with some of the main features of spatstat by working through a real data set of 3605 Beilschmiedia pendula trees in the tropical rainforest of Barro Colorado Island. This will give us a chance to learn what point process models are all about and how to fit them and also to learn about about R, by far the most commonly used statistical software today.

Feel free to post any questions or ideas here.

Jane Elith et al: range-shifting species (Friday 19th at 2)

The first paper in our Eco-Stats Symposium reading group is first-authored by the first speaker to present at the Symposium – Jane Elith.  It is called “The art of modelling range-shifting species”

http://onlinelibrary.wiley.com/doi/10.1111/j.2041-210X.2010.00036.x/abstract

and was published in Methods in Ecology & Evolution in 2010, the journal's first volume.  This paper has attracted a lot of attention, and is currently the second most highly cited paper in the journal, and by my estimation is solely responsible for about 11% of the journal’s 2011 impact factor!  We'll discuss this paper at UNSW on Friday 19th, 2-3pm and would be interested to hear what you make of it... feel free to post here any questions you have about it, comments, ideas...

Feel free to post any time, but if you aim for Friday 2-3 we might be able to get a discussion happening.