Normally Wed. evenings would be fine (almost any time is fine for me), but I
can't make it this month.
I had planned to briefly talk about two coursera classes that are in
progress now (about to start the 3rd week), both of which use R:
Computing for Data Analysis (4 wks)
https://class.coursera.org/compdata-2012-001/class/index
This class has been a useful review for me, and I've picked up a few R
tricks. I (re)discovered split followed by lapply/sapply -- and tapply can
also be quite useful. I didn't get much out of the first programming
assignment, but some new to R are struggling with it.
Mathematical Biostatistics Boot Camp (7 wks)
https://class.coursera.org/biostats-2012-001/class/index
This is a good refresher class for me, but some may be a bit overwhelmed by
the math. Calculus is regularly used in class lectures and is needed for
the quizes. R is occasionally mentioned. I'm not sure if there will be an
R programming assignment or not.
I thought this might be good class to recommmend to Stowers' scientists but
I think the math level may scare away many biologists.
There is a Data Analysis class that starts in Jan 2013 that might be useful
and will focus on the use of R. I plan to sign up for that.
https://www.coursera.org/#course/dataanalysis
I audited a few weeks of the Machine Learning class early this year and it
appeared to be an excellent class (but used Octave/Matlab instead of R):
https://class.coursera.org/ml/class/index
The one R trick I learned this month was that rownames/colnames should only
be applied to matrices and just happen to work for data.frames. For
data.frames one should use names for the columns and row.names for the rows.
I had a data.frame with 17 million rows and wanted to change one of the
column names. This took forever using colnames but only a few seconds with
names. I found this explanation online:
https://stat.ethz.ch/pipermail/r-devel/2006-September/042876.html. But I
rarely have 17 million rows <g>.
efg
Earl F Glynn
Overland Park