Enhancing Statistical Analysis with Markov Chain Models Using a Shiny R Interface

Torres-Cruz F.; Coaquira-Flores E.E.; Canqui-Flores B.; Ibañez-Quispe V.; Coyla-Idme L. · Lecture Notes in Networks and Systems (2024) · DOI: 10.1007/978-981-97-7423-4_36

Resumen

This study demonstrates the expansive utility of Markov chains in statistical modeling, emphasizing their role in simulating complex systems within diverse fields, including engineering, economics, biology, and computer science. We present an innovative integration of Markov chain theory to predict the future states of dynamic systems and introduce ModerCMarkov, a novel web application designed in R Studio using Shiny packages. This application leverages Markov chains for forecasting outcomes from varied database typologies. Our comprehensive evaluation of ModerCMarkov assesses its processing speed and predictive accuracy across multiple databases, varying in scope and complexity. The results highlight the application's robustness, evidenced by its rapid processing capabilities and precise predictions. Furthermore, our research utilizes the Markov chain approach to identify critical nodes within key variables, enhancing our understanding of these systems. ModerCMarkov emerges as a powerful tool for intricate analysis and modeling of complex variable databases, offering significant contributions to multidisciplinary research endeavors. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

Instituciones: Universidad Nacional del Altiplano