Modelling Internally Displaced Persons’ (IDPs) Time to Resuming their Ancestral Homes after IDPs’ Camps in Northern Uganda Using Parametric Methods

Abstract

In this paper the time that the Internally Displaced Persons (IDPs) took to return from IDPs Camps to their ancestral homes in Northern Uganda was modeled using parametric methods. A retrospective dataset was collected from a cohort of 590 households belonging to seven different villages from Otuke districts that were displaced by Lord Resistance Army (LRA) war and were used in the study for a period of seven years. The raw dataset shows a total of 66 households that have not yet return to their ancestral homes. Stata inbuilt program and Easyfit 5.5 professional software were used to test the distribution (Exponential, Weibull and Log-logistic) fitness for the retrospective IDP dataset. The three distributions fit statistic of Kolmogorov-Smirnov, Chi-Squared and Anderson-Darling were used to test for the distribution fit. The Weibull distribution model was found to have a superior fit for the data than both exponential and log-logistic distribution model since it had a wider acceptance region from the test statistics. The study however recommended the widening of the scope for data collection in future studies for better statistical inference.