Miners work injury determination using Bayesian structural equation model

Abstract

This paper develops a Bayesian structural equation model for miners work injury in an underground coal mine India. The technical and behavioural variables for work injury were identified and causal relationships were developed. The prior distributions of the causal parameters were obtained from the data obtained from the experts opinions and fitting the sample distribution with theoretical distribution by Chi-squared method. The posterior distributions of these parameters were obtained by applying Bayesian rule. The Markov Chain Monte Carlo simulation in the form of Gibbs sampling was performed for sampling the data from the posterior distribution. The results revealed that 19 parameters out of 33 causal parameters are statistically significant. The results of expert opinion based on priors and maximum likelihood priors revealed that the parameters bound were converged towards the expected value in spite of wrong or bad initialization of priors distributions. The error statistics reveals that Bayesian structural model provides reasonably good fit of work injury with high coefficient of determination (0.91) and less mean squared error (0.025).