Date of Award

May 2018

Degree Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Economics

Advisor(s)

William C. Horrace

Keywords

Econometrics, LASSO, Managerial Selectivity, Network, Productivity Analysis, Stochastic Frontier

Subject Categories

Social and Behavioral Sciences

Abstract

This dissertation develops empirical models that account for worker interactions, managerial selectivity, and technical inefficiency in the production process.

The first chapter, entitled "Stochastic Frontier Models with Network Selectivity," develops a model where workers produce output through peer-effect networks, while managerial selectivity of workers affects worker inefficiency. The intuition behind this model is that managers may consider optimal combinations of workers to produce the best results, and this selectivity in the worker network may affect worker productivity.

The second chapter, entitled "Network Competition and Team Chemistry in the NBA," models simultaneous interactions between multiple networks where agents cooperate with peers within their own networks but compete with non-peers from other networks. This paper presents the first econometric model to consider multiple peer networks where workers are engaged in simultaneous competition around a single outcome variable.

Lastly, the third chapter, entitled "Adaptive LASSO for Stochastic Frontier Models with Many Efficient Firms," develops a procedure to select a subset of maximally efficient firms in the sample of interest. In this model, firm inefficiency is measured as a distance from an estimated optimal production level, and I apply the LASSO (Least Absolute Shrinkage and Selection Operator, Tibshirani, 1996) to identify a subset of firms whose inefficiencies are estimated as exactly zero. This methodology can be applied to any classification problem where our interest is to identify a subset of best (worst) individuals among a large number of candidates.

Access

Open Access

Download

Share

COinS