Finetune intro

[ramhiser]

The CSDA reviewers made a few comments that lead me to believe the intro should be more clear. With that in mind, we need to finetune the intro and ensure we are communicating the following:

• We modify RDA to gain interpretability
• HDRDA inherently yields dimension reduction
• HDRDA is much faster than RDA
  • Mention savings in terms of Big Oh
  • Timing comparison (from #26)
• HDRDA should be used instead of RDA (for p > N?)
  • Add RDA to classification study?
• HDRDA is competitive in terms of classification performance

[ramhiser]

Maybe add a little more guidance around \alpha_k? Say something along the lines of, "We parameterize in such a way that it includes a few well-known models." If we do so, this will be need to be stressed in the intro as well as elaborated later in the paper.

[ramhiser]

Finally made progress on the intro. After input from @PhilYoung007, it's much stronger and gets to the point faster.

@PhilYoung007 you can find the updated version here: https://github.com/ramhiser/paper-hdrda/blob/revamp-intro/reports/paper-hdrda.Rnw

Click the Raw button to download the file.