[Statlist] Program of 4th Workshop "Applied Statistics in LifeSciences" - 15th April 2011, Bern

Tue Mar 29 20:04:21 CEST 2011

------------------------------------------------------
  4th Workshop "Applied Statistics in LifeSciences"

      15th April 2011, 09:45 – 12:30 (inclusive 30 min coffee break)
------------------------------------------------------
see also: www.imsv.unibe.ch/content/talks/sstats/spring_2011/index_eng.html

Location:  University of Bern, Institut für Exakte Wissenschaften,
           Room B78, Sidlerstrasse 5, 3012 Bern

Dear Colleagues

We are pleased to invite you to the 4th workshop "Applied Statistics in LifeSciences", 15 April 9 45 – 12 30h in Berne.

The three 15 min contributions with additional 15 min discussion time will be on (for more details of program and abstracts see below)
1) a dynamic model to analyse long-term changes in vegetation (Otto Wildi, WSL), 
2) a Bayesian Tobit model for precipitation (Fabio Sigrist, Seminar für Statistik, Zürich), 
3) a critical view on using the 1st principal component to do size correction in morphometrical analyses (Daniel Berner, Zoological Institute, Basel).

Please, send us a short note when you plan to participate, so that we can bring the appropriate amount of coffee and croissants. For people who announce their interest in a joint lunch until 10th April we will reserve a table in the Mensa or a nearby restaurant.

We are looking forward to a promising workshop and to seeing you there!

Fraenzi Korner and Barbara Hellriegel

Contact:
Dr. Fränzi Korner-Nievergelt, oikostat – Statistical Analyses & Consulting, <fraenzi.korner at oikostat.ch>
Prof. Barbara Hellriegel, Anthropology, University of Zurich, <barhell at aim.uzh.ch>

------------
Program
------------
10:15  Brief personal introduction of all Workshop participants 

10:25  Otto Wildi – Swiss Federal Institute for Forest, Snow and Landscape WSL, 
Birmensdorf 
Odyssey to a long awaited space-time process 

11:00  Fabio Sigrist – Seminar für Statistik, ETH Zürich 
A Bayesian Tobit model for precipitation 

11:35  Daniel Berner - Zoological Institute, University of Basel, 
Size correction in biology: how reliable are approaches based on (common) 
principal component analysis? 

12:10  Discussion: Communication between statisticians and users.  

12:30  Lunch  

-----------
Abstracts 
-----------

* A Bayesian Tobit model for precipitation 

Fabio Sigrist, Werner A. Stahel, Hans R. Künsch – Seminar für Statistik, ETH Zürich 
(sigrist at stat.math.ethz.ch) 

A spatio-temporal model for precipitation is presented. The model assumes that precipitation 
follows a censored and power-transformed normal distribution. Through a regression term, 
precipitation is linked to other covariates. Spatial and temporal dependencies are accounted 
for by a latent Gaussian variable that follows a Markovian temporal evolution combined with 
spatially correlated innovations. Such a specification allows for nonseparable covariances in 
space and time. Further, the Markovian structure yields computational efficiency and it 
exploits in a natural way the unidirectional flow of time. In addition, the model is space as 
well as time resolution consistent. The model is applied to 3 hourly Swiss rainfall data, 
collected at 26 stations. 

* Odyssey to a long awaited space-time process  

Otto Wildi – Swiss Federal Institute for Forest, Snow and Landscape WSL, 
Birmensdorf (otto.wildi at wsl.ch) 

Vegetation change is a space-time process, and a surprisingly slow one, as we know from 
many examples. Even if a survey is continued for decades it may just reveal an anecdote out 
of a long transformation sequence. I present an example from the Swiss National Park where 
a long-term project yields insight into almost 100 years of vegetation change. We used 
space-for-time substitution to uncover a temporal pattern at a scale of many centuries. A 
logistic competition model is composed yielding a straightforward explanation of the temporal 
patterns we encountered. However, to give a full account of the dynamic process 
geographical space needs to be included, allowing species to move as time passes. In my 
talk I shall demonstrate a modelled dynamic vegetation map of a 500 years succession and I 
summarize various missing links encountered in the course of analysis. 

* Size correction in biology: how reliable are approaches based on (common) principal 
component analysis? 

Daniel Berner - Zoological Institute, University of Basel, daniel.berner at unibas.ch 

Morphological traits typically scale with the overall body size of an organism. A meaningful 
comparison of trait values among individuals or populations that differ in size therefore 
requires size correction. A frequently applied size correction method involves subjecting the 
set of n morphological traits of interest to (common) principal component analysis [(C)PCA], 
and treating the first principal component [(C)PC1] as a latent size variable. The remaining 
variation (PC2-PCn) is considered size-independent and interpreted biologically. I here 
analyze simulated data and natural datasets to demonstrate that this (C)PCA-based size 
correction generates systematic statistical artifacts. Artifacts arise even when all traits are 
tightly correlated with overall size, and they are particularly strong when the magnitude of 
variance is heterogeneous among the traits, and when the traits under study are few. 
(C)PCA-based approaches are therefore inappropriate for size correction and should be 
abandoned in favor of methods using univariate general linear models with an adequate 
independent body size metric as covariate. As I demonstrate, (C)PC1 extracted from a 
subset of traits, not themselves subjected to size correction, can provide such a size metric. 

-----------------------
GROUP "LifeSciences" 
-----------------------
The GROUP "LifeSciences" within SSS serves as a networking instrument between 
practitioners and researchers in the Life Sciences interested in applied statistics and applied 
statisticians interested in life sciences topics. At the WORKSHOP, each participant will shortly 
introduce her/himself. Thereafter, three open statistical problems from the Life Sciences are 
presented (about 15 min), and discussed in plenum (about 15 min).  
On the one hand, this is a good opportunity for Life Sciences practitioners and researchers to 
get feedback and professional consulting on their data analyses. On the other hand, it 
presents problems and needs concerning applied statistics to the statisticians. This workshop 
promises synergistic gains for both sides. 

Contact:
Dr. Fränzi Korner-Nievergelt, oikostat – Statistical Analyses & Consulting, <fraenzi.korner at oikostat.ch>
Prof. Barbara Hellriegel, Anthropology, University of Zurich, <barhell at aim.uzh.ch>