AndeeKaplan

Assistant Professor at Colorado State University Statistics

Projects

Record linkage and the Downstream Task

Linking data from multiple databases increases both the size and scope of a dataset, enabling post-processing tasks such as linear regression or capture-recapture to be performed. This work focused on novel approaches to multiple downstream tasks that allow for errors to propagate through the analyses and provide accurate inference. Joint work with Brenda Betancourt and Rebecca Steorts.

[paper][package]

Computationally Scalable Bayesian Record Linkage

Linking data from multiple databases can increase the utility of many datasets and performing this linkage procedure using Bayesian methods can greatly enhance the analysis that results through the opportunity for error propagation. Unfortunately, Bayesian record linkage models are computationally complex and can be slow to fit. This work allows for greater scalability. Joint work with Neil Marchant, Rebecca Steorts, Ben Rubenstein, and Daniel N. Elazar.

[paper][github]

conclique

A fast way to simulate data with Markovian properties, which has provable MCMC geometric ergodicity. Implemented in a flexible R package that allows the user to specify an arbitrary model.

[paper][package]

Restricted Boltzmann Machines

Steps toward a thorough understanding of the RBM model class and its behavior from the perspective of statistical theory and exploration of the possibility of a rigorous fitting methodology via MCMC.

[paper][paper][shiny apps][github]

intRo

Shiny-based statistics learning application to foster student interest in coding while learning basic statistics. Joint work with Eric Hare.

[paper][tool][package]

Protoshiny

Shiny app to interactively visualize large dendrograms resulting from hierarchical clustering with prototypes. Joint work with Jacob Bien.

[paper][package]

gravicom

Graphical Visualization of Communities. A web application for community detection in network data through direct user interaction. Built on Shiny and D3.

[paper][tool][github]

NCS Dataviz

A graphical tool that allows users to understand variable relationships within the 2012 NCS Vanguard Study dataset. Built using D3. Joint work with Yongeng Lin from NORC.

[tool][github]

CommuniD3

Exploratory tool based on 'Soul of the Community' data generated by the Knight Foundation in cooperation with Gallup. Project was winner of the 2013 Data Exposition. Created using Shiny and D3. Joint work with Eric Hare.

[paper][tool][github]

Election 2012

Examined campaign contributions and political action committee spending data from the Federal Election Commission, and interpreted for the 2012 election cycle. Joint work with Di Cook, Heike Hofmann, Eric Hare, and Susan VanderPlas.

[paper][supplement][site][github]