Effects of Incorporating NETC School Enrollment Data in the Navy's Reenlistment Prediction (Roger) Model

摘要

The Navy's Selective Reenlistment Bonus Management System (SRBMS) uses a model known as ROGER to identify the SRB-eligible population and to predict the number of SRB takers for the following fiscal year. The Enlisted Bonus Manager uses the ROGER model to determine the SRB plans during the execution year. Over the years, constant changes in the structure of the SRB program have led to increasing levels of predictive error in the ROGER model. Specifically, the ROGER model has routinely under-identified the SRB-eligible population, which, in turn, led to under-predictions in the size of the predicted number of SRB takers and the SRB budget. The objective of this thesis is to determine whether the predictive errors in the Navy's SRBMS (ROGER) model can be reduced by accounting for new NEC/skill acquisition by sailors each fiscal year. NEC/skill acquisitions are accounted for by incorporating data into the ROGER model from the Naval Education and Training Command (NETC) on annual school enrollments and graduations. This thesis analyzes the impact of adding the NETC skill acquisition data to the ROGER model by analyzing the predicted SRB-eligible population and the predicted number of SRB takers and by assessing the resulting impact on the predicted SRB budget.