A Markov Decision Process in a Game of Studies
A Reinforcement Learning model for students’ investment decisions into higher education.
In this paper, we introduce a model for prospective students’ decision to join higher education and the optimal quantity to undertake. We set up a baseline model using a Markov Decision Process and depart from a perfect information case to include scenarios of omitted information about the agent’s reward, its transition probabilities and friction costs.
Data: National Longitudinal Survey of Youth Survey of 1979 (NLSY79)
Model: Markov Decision Process, Reinforcement Learning, Policy Iteration
Keywords: Markov Decision Process, Information Theory, Stochastic model, Transition matrix, Reward.
Software: RStudio, R (programming language)
Content Link Block
Select a page and create a visual link to it.
Learn more