Fifth Cross Validated Journal Club

Time to continue the CVJC series after holidays!

To remind for those who are new to the idea:

• CVJC is a whole day meeting on chat where we discuss some paper and its theoretical/practical surroundings.
• As mentioned above the event is whole-day (00:00-23:59UTC), but there is a meet-up session on which most talking take place.
• The paper must be OpenAccess or a (p)reprint suggested previously on a meta thread like this one and selected in voting.
• We will try to invite the author(s).

EDIT: The JC will certainly take place at 18.11.2011, with meet-up session at 18:00UTC. You can register here to get an e-mail reminder. One of the authors, Gregory Matthews, will join the discussion.

EDIT2: Here is the transcript.

This article was just posted on the Statistics Surveys website: Data confidentiality: A review of methods for statistical disclosure limitation and methods for assessing privacy.

This topic is becoming increasingly relevant and important.

A guide to modern statistical analysis of immunological data
Propose some questions to discuss:

• Typical types of distributions in biomedical (immunology, etc.) sciences
• Substantiation for transforming raw data with a given frequency distribution to those with a normal one

and some specific problems:

• Percentages in immunogram after angular (arc sin, $Freeman$ — $Tukey's$) transformation retain assymmetry. Only $Box$ — $Cox$ transformation sometimes helps well but power transformations for frequences (= percentages = proportions) is arbitrariness...
• Difficult asymetric, probably polymodal distributions (for example for cytokines) in some cases and almost bell-shaped ones in the others...