Add (optional?) bin-widths into the data files?

[manodeep]

The Baldry+2012 catalog contains the bin-widths as the second column. I am planning to use this catalog within an MCMC and therefore I would like to match the binning done in the data. However, since there is info about the bin-widths returned with the data, I can only guess as to what the widths were from diff'ing the x values. However, that produces two incorrect results:

• the number of bin-widths will be 1-less than the number of x values
• the correct bin-widths from the original catalog are 27 values like so [0.5, 0.5, 0.2, 0.2, ..., 0.2]. The diff produces 26 values of [0.5, 0.35, 0.2, 0.2, ..., 0.2]

I can handle this missing data within my workflow for this Baldry+2012 catalog but might be worthwhile to either implement/keep in mind for future use

[qyx268]

This is a good suggestion. I will keep it open for future development.

As I wrote this package back then for simply showing models against data, there are a number of things worthy of further improvement. However, as none of this type of observational data provides covariance between bins, these to me are just different levels of approximation when writing the likelihood :(