Plot/Statistics check

[SamHero16]

91cf671. Wanted to make sure these calculations and plotted statistics looked right, especially coverage and SE/SD.

Thanks.

[kpkeller]

I put a comment on #1 about some warnings I'm getting. But otherwise everything looks correct to me.

Here are some further suggestions/tips:

• I typically do not save pdf files to the git repo, since they can be quite large. As long as the source files are in the repo, anyone can recreate the pdfs. You should be able to "ignore" files in your git GUI (I forget right now which one you are using) so that they are not committed but stay out of the way.
• I would add something else to the file names to indicate which simulation they are from. This is less of an issue for this first demo simulation with a linear outcome, but for the Poisson one we will probably look at several different scenarios. Things to include could be like the grid size, a label for the confounder, or the residual error size.