August 2016 "Official BARUG" Meetup

Agenda

6:30PM - Appetizers and Networking
7:00 - Announcements
7:05 - Allan Miller: The rempreq R package
7:20 - Jocelyn Barker: Time Series Forecasting with R
8:00 - Nina Zumel: y-aware principal components regression in R

#####################################

Allan Miller

rempreq (https://github.com/amilleranalytics/rempreq): An R package for Estimating the Employment Impact of U.S. Domestic Industry Production and Imports

The U.S. Bureau of Labor Statistics publishes data which gives an indication of the relative impact of different industries production impact by year

The package rempreq includes both current and historic production data as well as functions to generate estimated time series for the employment impact for various types of production.

This presentation will include an introduction to rempreq, sample demonstrations of its use, and future plans for the extension of the package.

For a more details look here (http://schedule.user2016.org/event/7BXa/rempreq-an-r-package-for-estimating-the-employ).

################################

Jocelyn Barker

Time Series Forecasting with R

Time series forecasting has wide applications from inventory management and financial budgeting to predicting weather and the impacts of climate change. In this session we will cover the basics of time series analysis including nomenclature and common forecasting methods. We will discuss and demo R packages commonly used for forecasting including the “forecast” package. The talk will conclude by briefly covering the use of machine learning in forecasting including elements of experimental design unique to time series applications.

Bio:

Jocelyn Barker received her PhD in Biophysics from Stanford University in 2015 where she developed methods to use machine learning to diagnose medical images. She is currently employed by Microsoft where she forecasts revenue for Microsoft’s Central Finance Team. Her forecasts are directly consumed by Microsoft’s CFO, Amy Hood, who uses them to guide the revenue projections reported to Wall Street.

###############################

Nina Zumel

y-aware principal components regression in R

Y-aware techniques use the fact that for predictive modeling problems we know the dependent variable, outcome or y, so we can use this during data preparation in addition to using it during modeling.

Such methods include:

1. Effects based variable pruning

2. Significance based variable pruning

3. Effects based variable scaling

In this presentation Dr. Zumel will show how the incorporation of y-aware preparation into Principal Components Analyses can capture more of the problem structure in fewer variables.

To get a head start on the talk read her post on the subject (http://www.win-vector.com/blog/2016/05/pcr_part2_yaware/).