What to use as our StraightLine dataset?

[drphilmarshall]

Which is to say: what exercises do we want to do? And then, what's a good dataset to use? Suggestion so far: historical cepheids for H0! Assigning this to @elikrause to do with @beckermr but TBH we should all talk about this.

[drphilmarshall]

Hey @beckermr Can we use one or two of the datasets from our recent paper, please? Here's what we have to choose from:

I think I like 3982 for introductory purposes (its small, and has some obvious outliers), but I think we also want 4258 (for its very narrow straight line parameter likelihood). Can you please check these in to the examples/Cepheids folder as plain text files, with columns logP, M, err_logP, err_M, such that they can be loaded in with np.loadtxt, please? Thanks! :-)

@elikrause, do you have any other requests for Matt?

[beckermr]

Done! Sorry for the delay. I loaded up all of the data, but it is marked by galaxy number.

[drphilmarshall]

Coooool :-) Thanks Matt! The full dataset might come in handy for Week 3
Homework, in a hierarchical inference problem... or maybe in an example
project. For now, @elikrause and I will pick one galaxy out at a time as
required. I guess we could also concatenate them all together and enjoy the
resulting multi-modal posterior, eh @abmantz? Just a thought.

On Sat, Sep 12, 2015 at 10:09 PM, Matthew R. Becker <
notifications@github.com> wrote:

Done! Sorry for the delay. I loaded up all of the data, but it is
marked by galaxy number.

—
Reply to this email directly or view it on GitHub
https://github.com/drphilmarshall/StatisticalMethods/issues/8#issuecomment-139844908
.