MATSSforecasting is a research compendium for investigationg different approaches to forecasting ecological time series. It implements multiple methods to forecasting single time series, as well as metrics of time series complexity, with the goal of synthesizing the results to provide guidance on forecasting methods.
This project is set up as an R package compendium. What this means is much of the core functionality is bundled up into functions that are documented, much like an R package would. The recommended way to run the analysis, and/or contribute to the analyses is as follows:
There are two main ways to install the package. You can install it using the automated tools in
This will automatically install any dependencies, so can be a good way to start.
However, you will also want the analysis scripts, which are part of this github repo and not part of the package. You will want to clone this repo using Git. Here are some instructions if you are unfamiliar.
This will then enable you to get the most recent version of the code from within RStudio, by opening the project, clicking on the Git pane, and the “Pull” button.
Since this project is under active development, the codebase is likely to change rapidly from week to week.
With changes to the package components, you will want to re-build and install the latest copy. You can do this following the instructions above using
devtools. Or if you already have an updated copy within RStudio, use “Install and Restart” from the “Build” pane.