Selection of the best regression model to explain the variables that influence labor accident electrical company case

The present research proposes an alternative to select the best model that explains the relation of the variables that influence the labor accident in an electric power company. Among the techniques and tools used are those of occupational safety and health management, multivariate statistics, generalized linear models, the values of the deviation percentage explained and the adjusted percentage, and the Akaike and Bayesian information criteria. The following variables were identified through the mentioned techniques: management commitment, compliance with legislation, prevention planning, training in prevention, updating of occupational risk management and policies that have a significant influence on work accident and through The percentages and of the previously mentioned criteria were able to show that the logistic regression is the best model that explains the labor accident by presenting the highest percentage and the lowest values of the criteria when compared with the Poisson regression and negative binomial models.

• Autores: Varela Izquierdo, Noel; Viloria Silva, Amelec Jesus; Pérez Fernández, Damayse; Pineda Lezama, Omar Bonerge
• Tipo de recurso: Article of journal
• Fecha de publicación: 2017
• Institución: Corporación Universidad de la Costa
• Repositorio: REDICUC - Repositorio CUC
• Idioma: eng
• OAI Identifier: oai:repositorio.cuc.edu.co:11323/1767
• Acceso en línea: http://hdl.handle.net/11323/1767; https://repositorio.cuc.edu.co/
• Palabra clave: Hghest percentage; Information criteria; Multivariate statistics; Labor accident regression models; Negative binomial models
• Rights: openAccess
• License: Atribución – No comercial – Compartir igual