Comparison of ARIMA model, ARIMA-BPNN model and ARIMA-ERNN model in predicting incidence of dengue in China

## 1. OverviewThis repository contains all R scripts and supplementary materials used in the study to compare the predictive performance of three models:(1). ARIMA model (2). ARIMA-BPNN hybrid model (3). ARIMA-ERNN hybrid model All analyses were conducted using **R version 4.5.0** with the`fUnitRoots`, `tseries`, `forecast`, `astsa`, `urca`, `RSNNS`, `TTR`, `scales`, ` tidyverse `, `lubridate`, and `ggplot2` packages.## 2. DataThe monthly dengue incidence data (2004-2024) were obtained from the China Center for Disease Control and Prevention (China CDC).- The dataset includes 252 observations** (216 for training and 36 for testing).- The original dataset is shared publicly.- A sample dataset with the same structure (dengue.csv`) is provided for demonstration.- Variables: - `year`: month and year (e.g., 2004-01) - `incidence`: reported dengue incidence per month