This is the in-development version and major changes and corrections may be made - use at your own risk! Please share comments, suggestions and errors/bugs found, either directly on the GitHub page or by emailing [email protected].

NOTE: A major update was introduced on 21.03.2021. This should not affect results, but processing should be faster. It also removes the need in plotting functions to specify lots of details of model development, as this is now extracted automatically from the model (so arguments may need to be removed from existing code in order to run). As with any major change, there are likely to be bugs: bug reports are very welcome.

We are actively seeking review of the code - if you are able to provide feedback, we would love to hear from you (either on GitHub or at [email protected]).

epicoda is an R package designed to support epidemiological analyses using compositional exposure variables. It provides wrappers for common epidemiological use cases. Simulated data (simdata) can be used to try out the functions, and a vignette illustrates the steps to carrying out an epidemiological analysis with a Compositional Data Analysis approach to the exposure.

To install the epicoda package from GitHub:

install.packages("devtools") # To install epicoda from GitHub, the devtools package is required.  
library(devtools)
devtools::install_github("activityMonitoring/epicoda",  build_opts = c("--no-resave-data"), build_vignettes = TRUE, build_manual = TRUE)

epicoda can now be loaded as a normal package in R using:

library(epicoda)

To see examples of what the package can do, see the vignette (long form documentation with code and text). This uses an example analysis to illustrate how the package can be used. To view it, run:

vignette("vignette-epicoda")

This is the in-development version - please get in touch with any feedback or problems on this page, or by emailing [email protected]. We are aware of one issue where a conflict between dependency packages can lead to plots not displaying axis labels. The current settings should avoid this, but if it does affect you, it would be really useful to know.

If you use this package, please cite:

[Walmsley2020] Walmsley R, Chan S, et al. (2020)
Reallocating time from machine-learned sleep, sedentary behaviour or light 
physical activity to moderate-to-vigorous physical activity is associated with 
lower cardiovascular disease risk (preprint https://doi.org/10.1101/2020.11.10.20227769)