Online Program Home
My Program

Abstract Details

Activity Number: 232 - Methods and Tools for DoD Test and Evaluation
Type: Topic Contributed
Date/Time: Monday, July 30, 2018 : 2:00 PM to 3:50 PM
Sponsor: Section on Statistics in Defense and National Security
Abstract #329861
Title: CiTools: Quantifying Uncertainty for Statistical Models in R
Author(s): Matthew Avery*
Companies:
Keywords: confidence interval; prediction interval; R; ciTools; inference
Abstract:

This talk discusses tools for performing inference on predicted values in R with focus on random intercept mixed models.

ciTools is an R package that provides easy and intuitive methods for working with uncertainties for common types of model fits, including random intercept models. Existing tools in R tend to focus on parameter inference, leaving users who are interested in inference about model predictions (including confidence intervals, prediction intervals, quantile estimates and probability estimates) to build custom code. Different modeling tools (lm, glm, lmer, etc.) have unique interfaces, requiring users to learn different syntax for performing the same inference. ciTools provides simple functions for common types of statistical inference that have consistent sytanx and can be applied to different types of models.

Particular focus is placed on random intercept models fit with the lme4 package. ciTools provides users with a few reasonable approaches for this complex problem, allowing users to choose the best tool for their problem. We explore the properties of these intervals in a simulation study which is discussed in a tutorial that accompanies the package.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program