Change references to "Synthetic Data"

[ParadaCarleton]

The term "synthetic data" is ambiguous here, because in statistics, it usually refers to "fake data" created by fitting a generative model to a dataset, then drawing random data with the correct statistical properties from it. (Usually to preserve the privacy of participants while still providing a dataset for replication.)

[ablaom]

Thanks @ParadaCarleton for this.

Sure, within some Statistical community there might be some confusion. But looking around (see e.g. https://en.wikipedia.org/wiki/Synthetic_data) it seems a more general conception of "synthetic data" is pretty common.

How about we just add a clarifying sentence at the top of the "Generating Synthetic Data" section; something like

Here *synthetic data* means artificially generated data, with no reference to a "real world" data set. Not to be confused "fake data" obtained by resampling from a distribution fit to some  actual real data.

?

I think "example data" is too broad a term. This could be anything, real or imagined.

[ParadaCarleton]

Mostly I suggested this because when I tried to look up "Synthetic Data Generation Julia" or "Synthetic Population Julia" I kept getting this as a result 😅

[ablaom]

@ParadaCarleton Are you not happy with the smaller clarification I suggested? Your latest commit does not reflect it.

[codecov-commenter]

Codecov Report

Merging #1036 (75c5346) into dev (dd852d4) will not change coverage.
Report is 28 commits behind head on dev.
The diff coverage is n/a.

@@           Coverage Diff           @@
##              dev    #1036   +/-   ##
=======================================
  Coverage   60.97%   60.97%           
=======================================
  Files           2        2           
  Lines          41       41           
=======================================
  Hits           25       25           
  Misses         16       16           

see 1 file with indirect coverage changes

[ablaom]

Closed as rendered redundant by 2178c10