Accession Number : ADA496756
Title : Forecasting Marine Corps Enlisted Attrition Through Parametric Modeling
Descriptive Note : Master's thesis
Corporate Author : NAVAL POSTGRADUATE SCHOOL MONTEREY CA
Personal Author(s) : Hall, Jeremy T.
Full Text : http://www.dtic.mil/get-tr-doc/pdf?AD=ADA496756
Report Date : MAR 2009
Pagination or Media Count : 85
Abstract : The Marine Corps, as with any organization with a large workforce, must accurately monitor and more importantly predict the transition rates among personnel entering and exiting the enlisted and officer ranks. This emphasis is even more appropriate given that the Marine Corps has been authorized to increase the current authorized end strength by 13,000 personnel from Fiscal Year 2008 to Fiscal Year 2010. The purpose of this thesis is to apply parametric modeling (specifically survival analysis) to historical data sets of enlisted personnel in order develop a more efficient forecasting tool for military planners. It is the intent to include in the model those characteristics that significantly influence attrition behavior, and aggregate these findings to an efficient, yet effective forecasting model. Therefore, this thesis will analyze the interaction of time, individual characteristics, and those causal attributes that determine whether a Marine completes his or her contracted service. The current forecasting method used by the Marine Corps forecasts enlisted attrition annually. This study forecasts enlisted attrition monthly within occupational field. Hence, the data was structured to provide this depth of analysis. In comparison to the current forecasting method of exponential smoothing this study found that the use of survival analysis could be beneficial to not only forecast attrition, but also provide a descriptive assessment of attrition rates amongst occupation fields without loss of information due to averaging or weighting probabilities.
Descriptors : *PERSONNEL RETENTION , *FORECASTING , *MARINE CORPS , THESES , ATTRITION , PARAMETRIC ANALYSIS , MATHEMATICAL MODELS , ENLISTED PERSONNEL
Subject Categories : PERSONNEL MANAGEMENT AND LABOR RELATIONS
Distribution Statement : APPROVED FOR PUBLIC RELEASE