Patrick Kofod Mogensen defends his PhD thesis

Patrick Kofod Mogensen defends his PhD thesis:"Essays in Dynamic Economics"

Candidate

Patrick Kofod Mogensen

Title:"Essays in Dynamic Economics"

Time and place: 10 Marts 2021 at 10:00. Link to attend the defense follows here: https://ucph-ku.zoom.us/j/66802872236

An electronic copy of the thesis may be obtained here: charlotte.jespersen@econ.ku.dk

Assessment Committee

  • Associate Professor Thomas Høgholm Jørgensen, Department of Economics, University of Copenhagen, Denmark (chairman)
  • Professor Fedor Iskhakov, Australian National University
  • Associate Professor Fane Naja Groes, Copenhagen Business School


Abstract

This dissertation contains three self-contained chapters. They are united by the dynamic economic models that are either studied theoretically or applied empirically. The first two were written in collaboration with co-authors.

The first chapter is about the theoretical properties of the value function when solving discrete time, discrete choice dynamic programming problems using sieves to approximate the value function. The second chapter is about the incentives and dynamics that governs students’ progression and work choices. Using a dynamic structural model we explore behavior
of university students in Denmark and look into why students generally do not finish on time. We use the model to evaluate a number of counterfactual policies affecting university students. The third chapter derives equilibrium conditions for irectional dynamic games and shows how to solve them using homotopy continuation methods for systems of multivariate polynomials in the complete information formulation of the games and interval arithmetic for the incomplete information games.

Chapter 1 – Solving Dynamic Discrete Choice Models Using Smoothing and Sieve Methods with Dennis Kristensen, Jong-Myun Moon, and Bertel Schjerning (Forthcoming in Journal of Econometrics)

We propose to combine smoothing, simulations and sieve approximations to solve for either the integrated or expected value function in a general class of dynamic discrete choice (DDC) models. We use importance sampling to approximate the Bellman operators defining the two functions. The random Bellman operators, and therefore also the corresponding
solutions, are generally non-smooth which is undesirable. To circumvent this issue, we introduce smoothed versions of the random Bellman operators and solve for the corresponding smoothed value functions using sieve methods. We also show that one can avoid using sieves by generalizing and adapting the “self-approximating” method to our setting. We provide
an asymptotic theory for both approximate solution methods and show that they converge with square root-N-rate, where N is number of Monte Carlo draws, towards Gaussian processes. We examine their performance in practice through a set of numerical experiments and find that both methods perform well with the sieve method being particularly attractive in terms of computational speed and accuracy.

Chapter 2 – Student Choices, Incentives, and Labor Markets Outcomes: The Case of Delayed Graduation with Bjørn Bjørnsson Meyer. In this chapter, we set up a dynamic choice model describing how various pecuniary and nonpecuniary incentives influence university students’ decisions on part-time work, dropout, and delayed graduation. We estimate the model using Danish register micro data combined with administrative data from the country’s largest university. Counterfactual simulations using the estimated model show that: (i) About half of the average delay in time-to-graduation can be explained by students following economic incentives to prepare for the labor market with work experience. The other half is due to a range of factors, such as income through part-time work and grants and the cost of effort for heavy course load. (ii) Cutting financial aid with one year reduces average time-to-graduation by 0.3 year, but also increases dropout.

Chapter 3 – Equilibrium Conditions and Solution Methods for Directional Dynamic Oligopoly Games In this paper, I derive equilibrium conditions for sub-stages in directional dynamic games with different model specifications in terms of number of actions, number of players, and exogenous (non-)directional states. I show how to use these to solve for all Markov Perfect
Equilibria using Recursive Lexicographical Search. I add to the existing literature by deriving the needed equilibrium conditions needed to solve these games, and provide details on how to solve the sub-stages. I show how to solve the system of multivariate polynomial equations in complete information games using all-solution methods and propose a way to solve the more complex system of equations using interval arithmetic in incomplete information versions of some of the games. Full solution methods are important if the aim is to characterize the potential market configurations that can obtain, or if the goal is to estimate structural parameters in a model of dynamic, strategic interaction.