Hi All,
I have a straightforward question: does anybody know of a
peer-reviewed paper that used a competing risks model to examine the
effects of pesticides and parasites on bee colony mortality?
I need to check my references, but I think that Klein and Moeshberger
would be a commonly cited book for this type of analysis
http://www.amazon.com/Survival-Analysis-Techniques-Truncated-Statistics/dp/038795399X
and the original paper is
Gray, R. (1988), “A Class of K-Sample Tests for Comparing the
Cumulative Incidence of a Competing Risk,” The Annals of Statistics,
16, 1141–1154.
http://www2.math.uu.se/~garmo/Grey.pdf
There are many software implementations including
http://support.sas.com/resources/papers/proceedings12/344-2012.pdfhttp://www.jstatsoft.org/v38/i02/paper
You would look for a plot like the one in figure 1 of
http://www.jstatsoft.org/v38/i02/paper
that shows probability of death due to varroa compared to probability
of death due to pesticide exposure at some level.
There may be some issue since the Cox model doesn't support time
varying covariates but it's not clear that's a deal breaker here.
Another minor question I have in mind: In general, I'm interested in
understanding the differences in individual bee mortality and colony
mortality. (Is there a standard definition of colony mortality?) I
am interested in learning about any methods that people are using to
study the mortality of colonies in a formal statistical sense. I
would have expected to find a lot more papers using cox models or
models with time varying covariates. I did a quick search and have
only seen a couple papers that discuss this (in a methodology sense).
http://goo.gl/GA4zS2http://smallbluemarble.org.uk/wp-content/uploads/2012/04/029-034-VisserBlacquiere-2010.pdfhttp://www.ncbi.nlm.nih.gov/pubmed/14565579
Best,
Blair
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
