Added 'getting started' to documentation

[cgoldammer]

I followed the style in the DataFrames documentation. Thus, it's intended as a walk-through of the basic features without links to the more detailed documentation. For code blocks, the display differs from the rest of the documentation because I include the ">julia" from the REPL. This is because I want to include both code and output.

Very much looking forward to feedback. Happy to add more or do extensive editing as needed.

[johnmyleswhite]

I think using case only to distinguish an object from a type is a little confusing for a beginner.

[cgoldammer]

What would be easier? Omitting line 22 and just saying x=rand(Normal(), 100)? Or would you want me to use different notation?

[johnmyleswhite]

I often use d for a distribution object.

[johnmyleswhite]

This is great. Thanks so much for working on this.

[cgoldammer]

@johnmyleswhite, thanks for the comments, that all makes sense.

I edited based on your comments and did some general editing on top of it.

Also, I removed the Truncated type from the list of "main types", because that's likely confusing. Instead, I introduce the Truncated type through its own line. The main types are Univariate, Multivariate, and Matrixvariate.

[cgoldammer]

A quick note: Upon submitting my initial pull commit, the checks initially failed. Travis said that code coverage decreased by 10%. Within 1/2 hour or so, this got updated to the correct change (0%). For this commit, Travis fails to build (on the nightly build of Julia).

Not sure what's causing this issue. Happy to forward this issue to the right location if you know where that is.

[simonster]

This might be because we run code coverage on both Julia release and nightly and they give different results for reasons unrelated to the package.

[cgoldammer]

@simonster Good to know!. Apologies for closing and re-opening, just a bad click.

[simonbyrne]

Fantastic, thanks for doing this!