Development of a Markov Model for Forecasting Continuation Rates for Enlisted Prior Service and Non-Prior Service Personnel in the Selective Marine Corps Reserve (SMCR).

摘要

The purpose of this thesis is to develop a Markov model to determine the continuation rates for Prior Service and Non-Prior Service enlisted populations in the Selected Marine Corps Reserve (SMCR). Determining the end- strength for these populations is necessary for reserve manpower planners to balance the force structure to minimize personnel overages or underages that would impact training and labor costs as well as career progression. The results of model validation indicate that models based on annual aggregate monthly transition rates fail the stationarity assumption required of Markov models. These results suggest that attrition behaviors are seasonal for both enlisted populations, which leads to numerous states being nonstationary due to their correlation with seasonality. We recommend developing and employing models with unique transition rates for each month.