Ebi Antony George

Resources I have found useful

On this page, you will find links to online resources that I have found extremely useful. I'm putting them up here in the hope that they are of help to others too!

R

I taught myself R mainly from the internet, and these are the websites I kept going back to.

  • Introduction to R

    A basic introduction to R that covers a lot.

  • ggplot2

    The only package I use for plotting. It is very powerful and, quite frankly, I'm still learning what can be done with it, as it can do a LOT!

  • Top 50 ggplot2 Visualisations

    A useful list of many of the different visualisations that you can make.

  • cowplot

    My go-to package for publication-ready figures that combine multiple plots and images.

Statistics

My go-to statistical tool nowadays is mixed-effects modelling with an appropriate error structure. After years of trial and error, I now have a good workflow for these models, and the links below follow its steps.

  • Fitting distributions with fitdistrplus

    After looking at the data and some summary statistics, I use this package to compare how well the response variable fits various distributions.

  • GLMM FAQ

    An excellent overview, by Ben Bolker and others, of the R packages for fitting mixed-effects models. For simple linear and non-linear mixed-effects models I use lme4 or nlme.

  • gamlss

    One of the two packages I now use for more complex generalised linear models.

  • glmmTMB

    The other package I use for more complex generalised linear models.

  • DHARMa

    For testing model assumptions after fitting. Some packages have their own built-in functions, but I always use this one in addition.

  • emmeans

    An excellent package for post-hoc comparisons.

  • ggeffects

    For visualising model predictions, in addition to custom ggplot2 code.

Python

  • Python for Everybody

    The excellent set of five courses on Coursera from which I learned Python. It helped me find my way around the basics quite well.

  • PY4E

    Most of the material from that course can also be found on this website.

  • Using Python for Research

    A course on edX oriented more towards using Python in research. Its case studies are pretty great for introducing the numpy, matplotlib and pandas packages.

  • Using Pandas

    The pandas package makes handling data in Python quite similar to how it's done in R, and I cannot recommend it enough. This is one of the sites I keep going back to.

  • Pandas Cheat Sheet

    Another site I keep going back to, to get a hold on all that pandas can do.

Web development

I built the first version of this website from scratch, mainly with what I learned from the first two resources below.