On Mon, Oct 2, 2017, at 07:22 AM, Robert Horn wrote: > > Peter Davis writes: > > > Basically, I want to plot a time series graph showing my PSA (prostate > > specific antigen) over time. The PSA is measured at irregular intervals, > > and has been for over 4 years (and hopefully will continue for many more > > years.) That should be a simple enough graph. I've already got a > > javascript d3 example that does this, but I'd like to embed it in a > > document, and to be able to generate PDF. > > > > Further, I want to be able to show different time intervals with tinted > > bands spanning the full range of the graph, and having specific start > > and end dates. These would represent various medical treatments I've > > undergone. I have a rough example I've mocked up in Photoshop, but, of > > course, I want to be able to add new data and re-generate the chart as > > needed. I don't know if I can attach a PNG to an email on this list. > > > > I do something similar for managing diabetes. > > I use org-mode to manage some (not all) of the data tables and org-babel > to control a graphics and statistics analysis in R. R can also handle > input in other formats, such as CSV, that I get from some sources. The > results are also displayed in the org window as output from R. > > This is a much heavier weight solution, since it involves learning R. > But > the graphics capabilities are immensely richer than gnuplot and the > mathematical capabilities for statistics and time series analysis are > immensely richer in R. > > If learning R benefits your work or career you might explore this. >
Thanks, Robert. I've had considered R as a possible tool for this effort. I actually used it briefly for a data visualization course I took a few years ago, but there's still a lot more learning I would have to do to accomplish this. On the other hand, it seems every other path I've tried has a large learning curve too, so I should give R another look. Thanks, -pd -- Peter Davis www.techcurmudgeon.com
