dev-efficacy's Introduction

DEV-Efficacy

Code accompanying "Efficacy of dynamic eigenvalue in anticipating and distinguishing tipping points"

Notes on the files in this repository:

Always open DEV_Analysis.Rproj first before running the code from any file.
The files univariate_DEV.R and multivariate_DEV.R contain functions that evaluate univariate and multivariate DEV, respectively. These functions are imported and used by other files that generate the DEV results from time series.
The files discrete_model_functions.R, continuous_model_functions.R, non-smooth_model_functions.R and red_noise_functions.R contain functions that generate time series from the considered discrete-time models, continuous-time models, non-smooth (piecewise smooth) discrete models (all with additive white noise) and discrete models with multiplicative white noise, respectively. These functions are imported and used by other files that generate the DEV results from time series.
The file sensitivity_heatmap.R contains functions to generate a heatmap for sensitivity analysis. These functions are used by the file /plots/sensitivity_analysis/discrete_sensitivity_plots.R.
The folder data_generation contains the code to generate the data used for various plots. These data are saved in the data folder. The data have already been saved, and re-running the files in the data_generation folder would only overwrite them.
The folder plots contains files to generate various plots.
- /plots/avg_DEV/avg_DEV_discrete_plots.R: Fig. 1
- /plots/avg_DEV/avg_DEV_continuous_plots.R: Figs. 3, S1, S2
- /plots/avg_DEV/avg_DEV_non-smooth_plots.R: Figs. 4, S3
- /plots/rate_analysis/rate_anal_plots.R: Fig. 2
- /plots/red_noise/red_noise_plots.R: Fig. 5
- /plots/sensitivity_analysis/discrete_sensitivity_plots.R: Fig. 6, S4, S5