- #1

- 6

- 0

## Main Question or Discussion Point

I'm looking for a little crowd-sourced shepherding along the road to a more robust grasp of analytical methods for data munging and summarizing. I completed a BA in mathematics but focused on the pure thread, to the degree undergraduates can be said to focus on any one vein.

I was not a phenomenal student, and eschewed the applied courses as much as I could, so I'm not as strongly familiar with theoretical frameworks nor with cookbook-style production. That is, I know enough to know I don't know enough to deliver to my preferred level of quality as a data analyst. I can grapple with pivot tables pretty fluently, and I know we use arithmetic means as performance indicators without verifying they're appropriate. I am embarking on a basic analysis of the distribution of our data fields, and may begin presuming fit to various distribution models to see if I can fine-tune the analyses, but these amount almost to a random walk through the space of modelling techniques and undergirding theory.

I've tracked down Nelder's text, Generalized Linear Models (3rd Edition), but US$100 is a bit steep and I'm unsure how suitable it is for self-study. I also found Dobson's An Introduction to Generalized Linear Models which is both more affordable and by reviews seems a little gentler for the part-time, lone student. There are a bevy of others in Amazon's recommendations.

I am of course familiar with descriptive statistics, mostly mechanically, and completed a calculus-based statistics course which nevertheless required no integration of CDFs or any other evidence of actual calculus. I'm confident I can grok this stuff and am excited to learn more, both for my current employment but also for my own development. I've considered graduate work as a means of greater access to both the materials and mentors, but there are competing logistics I'd need to consider. Thus I'm looking to learn on my own.

Thanks for your time and brain power.

I was not a phenomenal student, and eschewed the applied courses as much as I could, so I'm not as strongly familiar with theoretical frameworks nor with cookbook-style production. That is, I know enough to know I don't know enough to deliver to my preferred level of quality as a data analyst. I can grapple with pivot tables pretty fluently, and I know we use arithmetic means as performance indicators without verifying they're appropriate. I am embarking on a basic analysis of the distribution of our data fields, and may begin presuming fit to various distribution models to see if I can fine-tune the analyses, but these amount almost to a random walk through the space of modelling techniques and undergirding theory.

I've tracked down Nelder's text, Generalized Linear Models (3rd Edition), but US$100 is a bit steep and I'm unsure how suitable it is for self-study. I also found Dobson's An Introduction to Generalized Linear Models which is both more affordable and by reviews seems a little gentler for the part-time, lone student. There are a bevy of others in Amazon's recommendations.

I am of course familiar with descriptive statistics, mostly mechanically, and completed a calculus-based statistics course which nevertheless required no integration of CDFs or any other evidence of actual calculus. I'm confident I can grok this stuff and am excited to learn more, both for my current employment but also for my own development. I've considered graduate work as a means of greater access to both the materials and mentors, but there are competing logistics I'd need to consider. Thus I'm looking to learn on my own.

Thanks for your time and brain power.