International Journal of Statistics and Systems

The empirical Bayes (EB) approach for parameter estimation has the capacity to yield more precise estimates than methods based on sampling theory. It can give sensible results in situations where sampling theory falters. It is this desirable property that has motivated this application in estimating indicators of vehicular traffic density (i.e. proportion of students who use personal motor vehicles) at the main campus of the University of Lagos (UNILAG for short). The EB estimator of a proportion P is the shrinkage estimator which is a weighted average of the prior proportion estimator _o and the sample proportion estimator _S. The EB estimators which we derived have lesser variance and very narrow confidence intervals than sample theory estimators. Hence, this article focuses on the application of an empirical Bayes (EB) model to the estimation of population proportions.