Analysing a randomised complete block design with vegan
03 November 2014 /posted in: R
It has been a long time coming. Vegan now has in-built, native ability to use restricted permutation designs when testing effects in constrained ordinations and in range of other methods. This new-found functionality comes courtesy of Jari (mainly) and my efforts to have vegan permutation routines use the permute package. Jari also cooked up a standard interface that we can use to drop this and some extra features neatly into any function we want; this allows us to have permutation tests run on many CPU cores in parallel, splitting the computational burden and reducing the run time of tests, and also a mechanism that allows users to pass a matrix of user-defined permutations to be used in tests. These new features are now fully working in the development version of vegan, which you can find on github, and which should be released to CRAN shortly. Ahead of the release, I’m preparing some examples to show off the new capabilities; first off I look at data from a randomized, complete block design experiment analysed using RDA & restricted permutations.
analogue 0.14-0 released
14 October 2014 /posted in: R
A couple of week’s ago I packaged up a new release of analogue, which is available from CRAN. Version 0.14-0 is a smaller update than the changes released in 0.12-0 and sees a continuation of the changes to dependencies to have packages in Imports rather than Depends. The main development of analogue now takes place on github and bugs and feature requests should be posted there. The Travis continuous integration system is used to automatically check the package as new code is checked in. There are several new functions and methods and a few bug fixes, the details of which are given below.
Simulating species abundance data with coenocliner
31 July 2014 /posted in: R
Coenoclines are, according to the Oxford Dictionary of Ecology (Allaby 1998), “gradients of communities (e.g. in a transect from the summit to the base of a hill), reflecting the changing importance, frequency, or other appropriate measure of different species populations”. In much ecological research, and that of related fields, data on these coenoclines are collected and analyzed in a variety of ways. When developing new statistical methods or when trying to understand the behaviour of existing methods, we often resort to simulating data with known pattern or structure and then torture whatever method is of interest with the simulated data to tease out how well methods work or where they breakdown. There’s a long history of using computers to simulate species abundance data along coenoclines but until recently no R packages were available that performed coenocline simulation. coenocliner was designed to fill this gap, and today, the package was released to CRAN.
Allaby M. et al. (1998) A Dictionary of Ecology, second. Oxford University Press.
Simultaneous confidence intervals for derivatives of splines in GAMs
16 June 2014 /posted in: R
Last time out I looked at one of the complications of time series modelling with smoothers; you have a non-linear trend which may be statistically significant but it may not be increasing or decreasing everywhere. How do we identify where in the series the data are changing? In that post I explained how we can use the first derivatives of the model splines for this purpose, and used the method of finite differences to estimate them. To assess statistical significance of the derivative (the rate of change) I relied upon asymptotic normality and the usual pointwise confidence interval. That interval is fine if looking at just one point on the spline (not of much practical use), but when considering more points at once we have a multiple comparisons issue. Instead, a simultaneous interval is required, and for that we need to revisit a technique I blogged about a few years ago; posterior simulation from the fitted GAM.
Identifying periods of change in time series with GAMs
15 May 2014 /posted in: R
In previous posts (here and here) I looked at how generalized additive models (GAMs) can be used to model non-linear trends in time series data. In my previous post I extended the modelling approach to deal with seasonal data where we model both the within year (seasonal) and between year (trend) variation with separate smooth functions. One of the complications of time series modelling with smoothers is how to summarize the fitted model; you have a non-linear trend which may be statistically significant but it may not be increasing or decreasing everywhere. How do we identify where in the series that the data are changing? That’s the topic of this post, in which I’ll use the method of finite differences to estimate the rate of change (slope) in the fitted smoother and, through some mgcv magic, use the information recorded in the fitted model to identify periods of statistically significant change in the time series.
Modelling seasonal data with GAMs
09 May 2014 /posted in: R
In previous posts (here and here) I have looked at how generalized additive models (GAMs) can be used to model non-linear trends in time series data. At the time a number of readers commented that they were interested in modelling data that had more than just a trend component; how do you model data collected throughout the year over many years with a GAM? In this post I will show one way that I have found particularly useful in my research.
File synchronisation with Unison
25 March 2014 /posted in: Computing
It’s becoming a fairly common experience to work on two or more computing devices; say a desktop/workstation in the office and a laptop when travelling or a home desktop. Which is great, but how do you keep all those machines in sync so that you have the latest versions of your files available no matter where you need to work?
Summarising multivariate palaeoenvironmental data part 2
09 January 2014 /posted in: R
The horseshoe effect is a well known and discussed issue with principal component analysis (PCA) (e.g. Goodall 1954; Swan 1970; Noy-Meir & Austin 1970). Similar geometric artefacts also affect correspondence analysis (CA). In part 1 of this series I looked at the implications of these “artefacts” for the recovery of temporal or single dominant gradients from multivariate palaeoecological data. In part 2, I introduce the topic of principal curves (Hastie & Stuetzle 1989).
Goodall D.W. et al. (1954) Objective methods for the classification of vegetation. III. An essay in the use of factor analysis. Australian Journal of Botany 2, 304–324.
Hastie T. & Stuetzle W. et al. (1989) Principal Curves. Journal of the American Statistical Association 84, 502–516.
Noy-Meir I. & Austin M.P. et al. (1970) Principal Component Ordination and Simulated Vegetational Data. Ecology 51, 551–552.
Swan J.M.A. et al. (1970) An Examination of Some Ordination Problems By Use of Simulated Vegetational Data. Ecology 51, 89–102.
Decluttering ordination plots part 4: orditkplot()
31 December 2013 /posted in: R
Earlier in this series I looked at the
ordilabel() and then the
orditorp() functions, and most recently the
ordipointlabel() function in the vegan package as means to improve labelling in ordination plots. In this, the fourth and final post in the series I take a look at
orditkplot(). If you’ve created ordination diagrams before or been following the previous posts in the irregular series, you’ll have an appreciation for the problems of drawing plots that look, well, good! Without hand editing the diagrams, there is little that even
ordipointlable() can do for you if you want a plot created automagically.
orditkplot() sits between the automated methods for decluttering ordination plots I’ve looked at previously and hand-editing in dedicated drawing software like Inkscape or Illustrator, and allows some level of tweaking the locations of labelled points within R.
Summarising multivariate palaeoenvironmental data part 1
28 December 2013 /posted in: R
Ordination methods that yield orthogonal axes of variation are often used to summarise the multivariate data obtained from sediment cores. Usually the first or, less often, the first few ordination axes are taken as directions of change or the main patterns of variance in the multivariate data. There is an oft-overlooked issue with this approach that has the potential to complicate the interpretation of the extracted axes, especially where there is a single or strong gradient in the data.