An official website of the United States government

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Bureau of Labor Statistics > Productivity Research and Program Development Group > Productivity research home

GASP 2019 presentations

Government Advances in Statistical Programming conference

September 23, 2019

Full GASP 2019 program
Videos from most GASP 2019 sessions are here on YouTube: https://www.youtube.com/@GASPFCSM

Session 1A: Using R to streamline analysis

Who is more likely to submit a 311 request? - Monica Puerto (American University)
Generic Drug Evaluation and R-package SABE, Elena Rantou (US Food and Drug Administration)
Using Amazon Web Services in R with the Paws package, David Kretch (Summit Consulting)
Using the Current Population Survey to Answer DOL-Relevant Questions: Applications in R, Samuel Rowe (Dept of Labor’s Office of the Assistant Secretary for Policy)

Session 1B: Model selection/validation with statistical software

A glimpse into the “Electricity Initiative” at the U.S. Energy Information Administration, Greg Lawson (Energy Information Administration)
Tableau for Data Scientists, Joel Hutchison (Tableau)
Using Cross-Validation for Variable Selection in Generalized Linear Models with Complex Survey Data, Darryl Creel (RTI International)
Scalable Approximate Kernelized Logistic Regression in R with an Application to the Census of Agriculture, Jonathan Abernethy (USDA-NASS)

Session 2A: Python libraries for text analysis and to streamline production

Easy deep learning text classification with gobbli, Jason Nance (RTI International)
Using Natural Language Processing and Machine Learning to Process Open Text Field Comments in a Panel Study, Kristin Chen (Westat)
Using Active Learning Framework and Automated Machine Learning Algorithms to Code Open-ended Responses, Rosalynn Yang (Westat)

Session 2B: Using R and open source to automate for efficiency

R Data Packages of Multi-Reader Multi-Case Studies and Simulation Tools to Support the Development of Reader Performance Evaluation Methods, Si Wen (FDA)
Hierarchical models in the production of official statistics: a discussion of some practical aspects, Andreea Erciulescu (Westat)
Topic Modeling Consumer Complaints for Risk Analysis, Benjamin Bloom (Federal Reserve Board)
Beyond the Analysis: An Approach to Operationalizing Your Results using Open Source JavaScript Frameworks, Dylan Holt (Accenture Federal)

Session 2C: Lightning Round!

Complex Survey Variance Estimation and Design Effects in R using the Rstan and Survey packages, Matt Williams (NCSES/NSF)
Ensemble Models with Basketball, John Thomas (George Washington University)
Automating the Maintenance of Survey Frames, Elizabeth Willhide and Keith Finlay (Census Bureau)
Text Analysis of Death Certificate Records to Ascertain Drugs Involved in Deaths from the National Vital Statistics System, Merianne Spencer and Brigham Bastian (NCHS)

Session 3A: Data dashboards for quick monitoring and analysis

SMART: An Open Source Tool to Facilitate Auto-Coding, Caroline Kery (RTI International)
Power Up NASS Data Analytics with MS Power BI, Leanne Tang (USDA-NASS)
Think Outside of the Box(plot), Jerry Valerio (Tableau)
Modern Techniques for Exploring Text Data, Peter Baumgartner (RTI International)

Session 3B: Open source software for fast information

Faster Computation for Hierarchical Bayesian-Model Using Rcpp Packages, Lu Chen (USDA-NASS)
Asking Consumers about their Finances, Kimberly Kreiss and Mike Zabek (Federal Reserve Board)
Using Python, PostgreSQL and R to analyze NIBRS data from the Crime Data Explorer, Ian Thomas (RTI International)
Weight Calibration across Packages, Dr. Stanislav Kolenikov-(Abt Associates)

Session 3C: Lightning Round!

Create Interactive Motion Charts Using R Package GoogleVis, Bidong Liu (Data and Analytic Solutions, Inc.) and Zhengyi Fang (Social & Scientific Systems, Inc.)
Automating Occupational Profiles Using RMarkdown, McLeod Brown (BLS)
Using R to Automate the Development of Custom Reports in Word Format, Alexandra Gates and Mina Zheng (NORC at the University of Chicago)
Linking EEOC Case Investigation Records by Employer Name using Text Analytics, Ada Harris (U.S. Equal Employment Opportunity Commission)
Rapid Implementation of Test Design Using Python, David Oh (BLS)

Session 4A: Application of Python and other software for production

Analyze, visualize, and . . . itemize: Tax policy analysis with Tax-Cruncher, Peter Metz (American Enterprise Institute)
Synergy between remote sensing and machine learning for crop yield prediction, Luca Sartore (USDA-NASS)
Getting Started with Using Amazon Mechanical Turk to Label Data, Brandon Kopp (BLS)

Session 4B: Analysis using R-shiny and other open source software

Using R and Shiny Applications to Analyze and Visualize NAEP Assessment Data, Emmanuel Sikali, Michael Lee, Paul Bailey and Ting Zhang (American Institutes for Research and NCES)
A Tour of USDA NASS’s Decision Support System, Nathan Cruze (USDA-NASS)
Measuring the cost and impact of open source software as intangible capital, Gizem Korkmaz (University of Virginia)

Session 4C: Lightning Round!

Adaptivity of Media and Behavioral Modeling, Mitchell Shuey (Independent) 
What is the meaning of this? How natural language processing can analyze unstructured data, Kelsey Gray (Insight Policy Research)
Routine Automation and Replication of PowerPoint Presentations in Python, Elaine WilcoxCook (Insight Policy Research) 
An Exploratory Research on Optimization of CFS Sampling Design, Mehdi Hashemipour (BTS), with Julie Parker
Science of Visual Analysis – Beyond Statistics, Jerry Valerio (Tableau)