Projects

Current Projects:

The projects listed here are the chapters of my disseration.

2020 Election Model:

I created a Bayesian hierarchical polling aggregation model that can run on a standard laptop in minutes while being almost as accurate in terms of average absolute error and mean square error as the FiveThirtyEight and Economist Model. Pre-print Coming soon. Google drive for election model.

Understanding Terrorism Policy Preferences through Bayesian Model Averaging and Multiple Imputation:

I applied Bayesian Model Averaging and Multiple Imputation to understand changes over time in a two-wave probability panel survey conducted in 2016 on Terrorism. I built regression models on the perceived likelihood of attacks, support for federal and local spending on terrorism, and support for new counterterrorism policies. Pre-print coming soon.

Model to Aggregate Two Public Opinion Panels:

I created a Bayesian model that integrates a two-wave probability panel and a six-wave non-probability panel on Terrorism policy preferences and perceptions to build regression models on the perceived likelihood of attacks, ,support for federal and local spending on terrorism, and support for new counterterrorism policies. Both surveys were conducted in 2016. These models help understand individual and population changes over time. The models are an extension of Multilevel Regression with Poststratification.

Former Projects:

A Series of Poll-Based Bayesian Models to Predict American Presidential Elections

The goal of this project was to study a variety of model specifications and compare the accuracy of these different approaches. A preregistered methodology is located here.

Predicting the 2016 Presidential Election:

I created a Bayesian model using the Gaussian conjugate prior. Poll data from other states served as the prior. Poll data from the state studied was used for the data. Predictions located here: http://2016usaelectionpredictions.blogspot.com/2016/11/final-call-summary.html and methodology here: http://2016usaelectionpredictions.blogspot.com/2016/09/looking-forward-to-november.html .

Right now I am exploring publication of the general election model.

A paper on the general election model was submitted to the Undergraduate Statistics Research Project competition sponsored by the American Statistics Assocation and the Consortium for the Advancement for Undergraduate Statistical Education. The article is located here.

A revised version of the article from the contest is now published in SIAM Undergraduate Research Online here.

Predicting Delegate Assignment in the 2016 GOP Nomination:

A simple bayesian model for delegate assignment in the 2016 GOP Nomination Process. Election results from states who already had a caucus or primary were used as the prior. Predictions and posts located here: 2016usaelectionprediction.blogspot.com .

Poster Presented at the Nebraska Conference for Undergraduate Women in Mathematics (2/5/17) and presented at the Texas Tech University Undergraduate Research Conference on 3/28.

Virtual Poster on General Election Model

Video presentation presented at Electronic Undergraduate Statistics Conference 2017.

Note: The video presentation contains a table that has the wrong values. A correct table and statement on the error can be found here.

Poster on the Second Set of Models Presented at the Conference of Texas Statisticians

This poster received 1st place in the Student Poster Competition at the Conference of Texas Statisticians.

Poster on the Second Set of Models Presented at the Texas Tech Undergraduate Research Conference 2018

You can contact me below about my research.

CV

Resume

Current Projects:

Former Projects:

Predicting the 2016 Presidential Election:

Predicting Delegate Assignment in the 2016 GOP Nomination:

Graduate Presentations

Graduate Posters

Undergraduate Presentations

Undergraduate Posters