schamberlain.github.io

fromthebottomoftheheap.net

Opportunities to join The Lab

http://www.fromthebottomoftheheap.net/lab/join

From the bottom of the heap. Opportunities to join The Lab. Currently there are no open positions in the lab, but I’m always interested in talking with potential graduate students or postdoctoral fellows about opportunities for funding and research ideas, so do please get in touch. Andrew Barr's Ancient Eco. Methods in Ecology and Evolution. Musings on Quantitative Palaeoecology. Used under CC-BY licence.

fromthebottomoftheheap.net

R scripts for DPER volume 5

http://www.fromthebottomoftheheap.net/code/dper-scripts

From the bottom of the heap. R scripts for DPER volume 5 data handling methods and numerical techniques. I wrote or co-wrote 3 chapters in the recently-released 5th volume in the Developments in Palaeoenvironmental Research series. Tracking Environmental Change Using Lake Sediments: Data Handling Methods and Numerical Techniques. Statictical Learning in Palaeolimnology (Chapter 9; with John Birks),. Analogue Methods in Palaeolimnology (Chapter 15), and. Statistical Learning in Palaeolimnology.

fromthebottomoftheheap.net

Statistical Learning in Palaeolimnology

http://www.fromthebottomoftheheap.net/code/dper-scripts/chapter-9-statistical-learning.html

From the bottom of the heap. Statistical Learning in Palaeolimnology R script for examples. Whilst I’m preparing an annotated set of scripts for the Statistical Learning chapter of DPER5, you can access the updated R code used for the chapter via my dper5. The following scripts cover the examples from the chapter. Multivariate adaptive regression splines (MARS). Shrinkage methods in regression (ridge, LASSO, elastic net). Script requires the development version of my analogue. Andrew Barr's Ancient Eco.

fromthebottomoftheheap.net

Members of the lab

http://www.fromthebottomoftheheap.net/lab/members

From the bottom of the heap. Members of the lab. Current members of the lab are:. Dr Emma Wiik (Postdoctoral Fellow). Emma joined the lab in March 2015 and is a joint appointment between my lab and Peter Leavitt’s Limnology Laboratory at the University of Regina. Emma’s research in the lab focuses on the the effects of climate variability on pairie lakes, with both palaeo- and neo-limnology aspects. Brittany Hesjedal (MSc Candidate). Andrew Barr's Ancient Eco. Methods in Ecology and Evolution.

fromthebottomoftheheap.net

cocorresp

http://www.fromthebottomoftheheap.net/code/r-packages/cocorresp.html

From the bottom of the heap. Cocorresp an R implementation of co-correspondence analysis. CoCA’s response model is the same as that of correspondence analysis (CA) and hence is particularly suited to species abundance data. Symmetric CoCA is closely related to the technique of coinertia analysis, whilst predictive CoCA employs partial least squares to handle the case of a large number of predictor variables. Aim to emulate vegan. Functions but for CoCA ordinations. Bugs, feature requests.

fromthebottomoftheheap.net

The new Tri-agency open access policy

http://www.fromthebottomoftheheap.net/2015/07/10/tri-agency-open-access-policy

From the bottom of the heap. ESA's publishing deal with Wiley. My aversion to pipes. The new Tri-agency open access policy. In the main, the Tri-Agency policy on open access. Is pretty simple and to be fair to the Tri-Agencies, they have done a good job of providing information and FAQs. But first, a summary of what the policy actually requires:. Within 12 month of publication the peer-reviewed, final author-version of the manuscript. Research grant funding from one of the Tri-Agencies awarded May 1st 20...

datavoreconsulting.com

Count data and GLMs: choosing among Poisson, negative binomial, and zero-inflated models | Datavore Consulting

http://datavoreconsulting.com/programming-tips/count-data-glms-choosing-poisson-negative-binomial-zero-inflated-poisson

Math, programming, ecology, and more. Count data and GLMs: choosing among Poisson, negative binomial, and zero-inflated models. Ecologists commonly collect data representing counts of organisms. Generalized linear models (GLMs) provide a powerful tool for analyzing count data. [As mentioned previously. You should generally not transform your data to fit a linear model and, particularly, do not log-transform count data. No of samples per treatment. Mean count for treatment A. Mean count for treatment B.

datavoreconsulting.com

Numerically solving PDEs in Mathematica using finite difference methods | Datavore Consulting

http://datavoreconsulting.com/programming-tips/numerically-solving-pdes-mathematica-finite-differences

Math, programming, ecology, and more. Numerically solving PDEs in Mathematica using finite difference methods. Sadly, some types of PDEs are beyond NDSolve’s capabilities. Confronted with one of these PDEs, a user must resort to a more “manual” procedure to find a numerical solution. In this post, we’ll examine a few tricks that can make this process easier. Consider the following PDE:. We seek a solution,. Xgrid = Range[0, 10]; ygrid = Range[0, 10]; grid = Outer[{#1, #2} &;, xgrid, ygrid];. At each grid...

datavoreconsulting.com

On Labor Day, make your computer’s job easier with Milstein’s method | Datavore Consulting

http://datavoreconsulting.com/programming-tips/labor-day-computers-job-easier-milstiens-method

Math, programming, ecology, and more. On Labor Day, make your computer’s job easier with Milstein’s method. In today’s post, we will explore numerical schemes for integrating stochastic differential equations in Mathematica. We will take an informal approach; for an in-depth treatment of stochastic differential equations, I recommend that you look at. Stochastic Processes for Physicists. Modeling with Ito Stochastic Differential Equations. In the realm of stochastic differential equations, there are much...

fromthebottomoftheheap.net

Permissions

http://www.fromthebottomoftheheap.net/permissions

From the bottom of the heap. Unless stated otherwise From the bottom of the heap. By Gavin L. Simpson. Is licensed under a Creative Commons Attribution 3.0 Unported License. This means you are free to re-use content on my blog however you might wish to, so long as you attribute the source to me, Gavin L. Simpson. There are a couple of exceptions to this where I have used the work of others. Unless stated otherwise, and where applicable, software and code on this blog are both released under the GNU.