Fitting the distributions of length of stay in Sultan Qaboos University hospital using transformed distributions

Fitting real life data using probability densities has a long history of development. The purpose of this study is to assess the appropriateness of new constructed probability densities for fitting the empirical distribution of the length of stay (LOS) data, which are collected from the Intensive Care Unit (ICU) at Sultan Qaboos University Hospital (SQUH). Analysis of LOS data using well known distributions, such as Log-normal, Weibull and Gamma, were widely considered in the literature over the last three decades. Introducing new distributions with more flexibility may result on significant improvements of the existing findings. An empirical comparative study will be conducted. Indeed, models providing better fits could be used for forecasting the length of stay of patients in the hospital; and, consequently, could help the decision makers of SQUH in terms of budget reduction.