Pivot-Madagascar/PRIDEC-package

R disease forecasting package for the PRIDE-C project # Predicting Infectious Diseases via Environment and Climate (PRIDE-C) The goal of PRIDE-C is to provide a standardized API functionality for forecasting infectious diseases from DHIS2 data. ## Installation You can install the development version of PRIDEC from GitHub with: ``` r # install.packages("devtools") devtools::install_github("Pivot-Madagascar/PRIDEC-package") ``` ## Example TThe PRIDE-C forecasting approach follows the following steps: 1. Data processing 2. Model tuning and training 3. Forecasting The example below follows the steps for fitting a Random Forest model using `ranger` from simulated model data. ``` r library(PRIDEC) data(demo_malaria) #preprocess data set data_clean Registered S3 method overwritten by 'quantmod': #> method from #> as.zoo.data.frame zoo #create a cv_fold of assessment (historical) and analysis (forecast) data cv_set Registered S3 method overwritten by 'scoringutils': #> method from #> print.forecast forecast #> # A tibble: 2 × 10 #> dataset wis mae med_ae mean_ae_log wape dispersion sp_rho prop_over #> #> 1 analysis 5.82 6.48 0.190 0.0117 0.151 5.13 0.965 0.200 #> 2 assess 24.4 33.1 32.0 1.09 0.605 2.33 0.808 0.0317 #> # ℹ 1 more variable: prop_under plot_predictions(rf_fit[rf_fit$orgUnit %in% sample(rf_fit$orgUnit,1),]) ``` ## Contribute to PRIDE-C As an open-source package, we welcome all contributions. Please feel free to file an issue or contact the developer (@mvevans89). ## Funding The development and maintenance of this package is funded by a Wellcome Trust Digital Technology Development Award.