Minor clarifications in the TSDiD vignette

[Ph-ADD]

• added some details about traditional TWFE
• added markdown notes / warnings for key points
• misc clarifications about the did2s method / functions

[Ph-ADD]

Just my attempt at clarifying some things in the vignette while teaching myself the method.

The technical changes are minor but should be reviewed. The main technical change was on the subscripts of the first TWFE equation. The first sum ended at -2 and the second sum started at 1. As far as I can tell, it's usually fine to estimate the treatment period (t = 0). Just that one period needs to be dropped to avoid multi-collinearity. But I'd be happy to learn why I'm wrong on this.

The rest of the technical description changes were about the traditional TWFE model / notation.

[kylebutts]

All of this is helpful and appreciated!

Yes, we do not need to drop -1 for did2s. That was a mistake that I thought I fixed in a previous iteration. If you could change that too, that would be great!