Web API of the Central Bank of Honduras: Evaluation of Its Integration in R and Perspectives for the Development of a Specialized Package

The automation of economic analysis through open data interfaces has become a strategic component in the development of modern data infrastructures. This study evaluates the integration and analytical potential of the public Web API of the Central Bank of Honduras (BCH), focusing on its applicability within the R programming environment. Through a bibliometric and systematic literature review, global trends in the use of APIs for economic, computational, and predictive modeling purposes are identified. A case study is presented using the year-on-year variation of the Consumer Price Index (CPI) as a proxy for inflation, extracted directly from the BCH API. Six time series forecasting models were implemented—ARIMA, Simple and Linear Exponential Smoothing, Prophet, Random Forest, and XGBoost—comparing their performance using MAE, RMSE, and MAPE metrics. The results show that machine learning models outperform traditional methods in capturing nonlinear patterns in inflation dynamics. This research highlights the relevance of R as a comprehensive analytical environment for data acquisition, processing, modeling, and visualization. Additionally, it underscores the strategic value of programmable access to official statistics and proposes the development of a dedicated R package for the BCH API as a means to enhance reproducibility, efficiency, and transparency in economic research and public policy design.