Hello, my name is Ron Wehrens.
I am a researcher at Wageningen University and Research at the Biometrics Department.
And today I'm going to talk about "Multivariate Statistics in the Context of Metabolomics".
So, whenever you do data analysis,
it's really important to look at your data and
to visualize them in several different ways.
This is also discussed in the lecture on univariate statistics.
And one of the reasons for doing this is that
the human mind is a very good pattern recognizer,
so we are able to see patterns that are very hard to pick up by automated methods.
So things that we might pick up are outliers,
or we might see relationships between variables.
And we might also assess whether the plots that
we see confirm our expectations on the data.
So, one of the key elements of doing
graphics is trying to visualize information in each graph.
And you can do that in a good way and in a bad way.
A very good reference on making good graphics is the book by Bill Cleveland,
already more than 20 years old, "Visualizing Data".
And I would recommend anyone to pick up a copy at the library and look at it.
R is one of
the most popular data analysis programs
currently around and it is also the program that we are using in this lecture series.
This is a small exercise,
showing you the power of some of
the built-in visualization tools that are there in R. So,
you can simply copy and paste the commands that are on
this slide and take a break from this lecture,
do the R session, do the examples,
and when you're ready, come back to this lecture,
and we can proceed.