Degree Type

Dissertation

Date of Award

2016

Degree Name

Doctor of Philosophy

Department

Statistics

Major

Statistics

First Advisor

Dan Nettleton

Abstract

Within this dissertation are 3 papers application of statistical analyses to data in sport. We discuss the common methods of estimating in-game win probability values and present an approach using random forests that is uniformly applicable to all head-to-head competitions. The random forest is a non-parametric machine learning methodology common in big data regression and classification problems. We demonstrate the performance and usefulness of our method to the NHL, NBA and NFL. We also introduce a new methodology to account for missing values that are associated with the linear predictor in order to improve the estimation of NFL field goal kicker accuracy. Due to its flexibility, we believe the that the framework for incorporating information underlying missing values could be useful in a wide array of applications.

DOI

https://doi.org/10.31274/etd-180810-5589

Copyright Owner

Dennis Lock

Language

en

File Format

application/pdf

File Size

70 pages

Share

COinS