---------------------------------------
Statistics seminar
School of Mathematics
The University of Edinburgh
---------------------------------------
Friday 1st March, 3.10 pm Room 5323, JCMB
Speaker: Javier Palarea-Albaladejo,
BioMathematics and Statistics Scotland
Handling values below detection limit in compositional data sets
Abstract
Samples representing part of a whole, usually called compositional or
closed data in statistics, are common in many fields of the earth and
life sciences—say chemical concentrations, food compositions,
activity/behavioural patterns, abundance of species, and so on. Their
distinctive feature is that there is an inherent relationship between
all the parts constituting a sample, as they only convey relative
information. On the other hand, one often finds that e.g. some analytes
are not present in sufficient concentration to allow the measuring
instruments to effectively detect them. These nondetects are usually
labelled as “<DL” (less-thans), indicating that the actual values are
below known limits of detection. Many data analysis techniques require
complete data sets, so there is a need of sensible replacement
strategies for less-thans. The peculiar nature of compositional data
determines any data analysis and demands for specialised treatment of
less-thans. Some basic principles of compositional data analysis and
well-founded statistical methods for nondetects will be presented aiming
to prevent practitioners from relying on popular but untrustworthy
approaches.
This seminar is joint with BioSS and is a part of Maxwell Institute
Statistics seminar.
There will be tea and coffee in the Mathematics Common Room (5212) after
the seminar.
The seminar websites:
http://www.maths.ed.ac.uk/events/statistics (in google calendar)
http://www.maths.ed.ac.uk/~nbochkin/StatisticsSeminar.html
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|