I have noticed that whenever I look at the actual numbers associated with a
time series or some calculation from the data (a single trial average, for
instance) they are all integers. Even if I use 3dcalc to make sure I've
created a fraction, I still get an integer when I look at the graph of the
new anatomy dataset I've created.
Why is this? Is this a function of the way AFNI does things, or is this
just something that inevitably goes with fMRI data (something to do with
the "arbitrary units" associated with BOLD signal)?
If it's something within AFNI, what is the rounding algorithm used?
Won't rounding compromise my statistics at some level (I mean, allowing
only integers, instead of fractions, must change the variability in one way
or the other)?

Thanks.

------------------------------------------
Posted to Phorum via PhorumMail