This post expands on Part 3 -- specifically the
section concerning Outcome Approximations -- and begins preparing data for
analysis.
In this post, I will compare the estimation method used in the paper vs an
alternative, and then look at a few other published methods. The estimation is
for the
outcome …
This post continues from Part 3 and covers the
Bayesian profile regression model.
Bayesian Profile Regression
What Is It?
There were several resources I used to get a grasp on the method. Molitor et
al. 2 present kind of the founding proposal of the method, and this is a …
This post continues from Part 1 and covers the
standardization of the
covariates used for the
clustering part of their model.
I think it'll wind up pretty short. I had initially created some maps to
show off, but decided that they didn't add anything substantial to the
explanation.
Wamsley M, Coker ES, Wilson RT, Henry K, Murphy HM (2024) Social
vulnerability and exposure to private well water. PLOS Water 3(12): e0000303.
https://doi.org/10.1371/journal.pwat.0000303
In the previous post on sand crabs, I
looked at the mortality of the treatments (exposed to environmentally
relevant concentrations of microplastic
fibers)
vs controls. Now, I want to look at another outcome that they looked at: egg
development.
Specifically, looking at starting vs ending egg development stages. This …
This post should hopefully be helpful for people like me who aren't
statisticians, but can work with R decently.
While working with count data, I encountered a new type of residual
(apparently there's different types)
called the "randomized quantile residual" (RQR).
RQRs essentially convert your residuals such that you …