Publications by ucfagls
Smoothing temporally correlated data
Something I have been doing a lot of work with recently are time series data, to which I have been fitting additive models to describe trends and other features of the data. When modelling temporally dependent data, we often need to adjust our fitted models to account for the lack of independence in the model residuals. When smoothing such data, ...
5044 sym R (1230 sym/4 pcs) 42 img
Passing non-graphical parameters to graphical functions using …
Argument passing via ... is a great feature of the R language, allowing you to write wrappers around existing functions that do not need to list all the arguments of the wrapped function. ... is used extensively in S3 methods and in passing graphical parameters on to graphical functions. When writing you own plot methods, using ... allows the use...
3090 sym R (1908 sym/8 pcs) 16 img
permute: a package for generating restricted permutations
Multivariate ordination methods are commonly used in ecology to investigate patterns in species composition in space or time. Constrained ordination methods such as redundancy analysis (RDA) and canonical correspondence analysis (CCA) are effectively just multiple regressions, but we lack the parametric theory to adequately test the statistical s...
7520 sym R (1208 sym/7 pcs) 18 img
Generating restricted permutations with permute
In a previous post I introduced the permute package and the function shuffle(). In that post I got as far as replicating R’s base function sample(). Here I’ll briefly outline how shuffle() can be used to generate restricted permutations. shuffle() has two arguments: i) n, the number of observations in the data set to be permuted, and ii) con...
3609 sym R (2260 sym/6 pcs) 16 img
Generating sets of permutations
In previous posts I discussed how to generate a single permutation from a fully-randomised or restricted permutation design using shuffle(). Here I want to briefly mention the shuffleSet() function and illustrate it’s usage. Every time you call shuffle() it has to interpret the control list to identify the type of permutation required. Whil...
3836 sym R (2596 sym/6 pcs) 16 img
New version of analogue (0.8-0)
Yesterday I pushed an update of my analogue package to CRAN. The new version is 0.8-0 and contains some new functions, several bug fixes and a major change arising from additions to R 2.14.x requiring all packages to have a namespace. analogue now has its own namespace rather than relying on the one R would automagically generate if it weren’t ...
2812 sym Python (3155 sym/1 pcs) 16 img
A better way of saving and loading objects in R
Hadley Wickham (@hadleywickham) this week mentioned on Twitter his preference for saveRDS() over the more familiar save(). Being a new function to me, I thought I’d take a look… save() and load() will be familiar to many R users. They allow you to save a named R object to a file or other connection and restore that object again. When loaded ...
2760 sym R (588 sym/2 pcs) 16 img
Course: Analysing Palaeolimnological Data with R
Update: As of 18 April 2012 the course is now fully subscribed. Should you wish to be placed on a waiting list please contact Steve Juggins. Analysing Palaeolimnological Data with R 16th — 20th August 2012, University Marine Biological Station Millport, Isle of Cumbrae, Scotland Steve Juggins and I will be running a 4-day residential R course t...
2223 sym 4 img
Customising vegan’s ordination plots
As a developer on the vegan package for R, one of the most FAQs is how to customise ordination diagrams, usually to colour the sample points according to an external grouping variable. Now, just because we get asked how to do this a lot is not really a reflection on the quality of the plot() methods available in vegan. Ordination diagrams are dif...
7229 sym R (2891 sym/17 pcs) 10 img
Quantitative palaeolimnology: my book chapters are finally out!
Today I received confirmation that the delayed fifth volume in the Developments in Palaeoenvironmental Research series has been published. The book is titled Data Handling and Numerical methods, though it covers more of the latter and, IMHO, is far more interesting than the dry title would suggest (who gets excited by Data Handling? Well, one o...
3071 sym 6 img