Online Program Home
  My Program

All Times EDT

Abstract Details

Activity Number: 27 - SDNS Speed Session
Type: Contributed
Date/Time: Sunday, August 8, 2021 : 1:30 PM to 3:20 PM
Sponsor: Section on Statistics in Defense and National Security
Abstract #318696
Title: Spatiotemporal Modeling of Career Pathways
Author(s): Aritra Halder* and Joshua Randall Goldstein and Joanna Schroeder and Sallie Keller
Companies: University of Virginia, Biocomplexity Institute and Biocomplexity Institute, University of Virginia and University of Virginia, Biocomplexity Institute and Initiative and Biocomplexity Institute, University of Virginia
Keywords: career pathways; Markov decision processes; Bayesian multinomial logistic; spatial and temporal variation
Abstract:

Organizations are generally structured hierarchically such that employed individuals with increasing knowledge, skills, and abilities are able to navigate between positions. From the perspective of the organization, this produces career pathways and trajectories for employed individuals subject to various professional constraints which require modeling. An ideal predictive framework to model these transitions should consider individual characteristics and quantify unobserved risks that the employee faces from various sources. In this work, we construct such a framework for Army veterans with the aim of modeling their career trajectories on exiting military service, using resume data provided by Burning Glass Technologies. The time horizon considered for modeled pathways is 10 years. The spatial domain is restricted to include the District of Columbia, the state of Maryland, and Virginia. We propose a Bayesian multinomial logistic model for transitions between states, while accounting for risks conferred from spatial and temporal sources. We present extensive synthetic experiments documenting the efficacy of our algorithm and compare results to existing modeling frameworks.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2021 program