Analyzing positive data with 0 values can be challenging, since a direct log transformation isn't possible. I discuss some of the things to consider when deciding on an analysis strategy for such data and then explore the effect of the value of the constant, c, when using log(y + c) as the response variable.
In this post I show an example of how to automate the process of making many exploratory plots in ggplot2 with multiple continuous response and explanatory variables. To loop through both x and y variables involves nested looping. In the latter section of the post I go over options for saving the resulting plots, either together in a single document, separately, or by creating combined plots prior to saving.
Checking for autocorrelation must be done carefully when some observations are missing from a time series or the time series is measured for independent groups. I show an approach where I pad the dataset with NA via tidyr::complete() to fill in any missed sampling times and make sure groups are considered independent prior to calculating the autocorrelation function.
Unstandardizing coefficients in order to interpret them on the original scale can be needed when explanatory variables were standardized to help with model convergence when fitting generalized linear mixed models. Here I show one approach to unstandardizing for a generalized linear mixed model fit with lme4.
Checking for model fit from generalized linear mixed models (GLMM) can be challenging. The DHARMa package helps with this by giving simulated residuals but doesn't work with all model types. I show how to use tools in DHARMa to extend it for use with unsupported models fit with glmmTMB() and zeroinfl().