Online Program
Program-at-a-Glance
Keynote Address | Concurrent Sessions | Poster Sessions
Short Courses (full day) | Short Courses (half day) | Tutorials | Practical Computing Demonstrations | Closing General Session with Refreshments
Keynote Address | Concurrent Sessions | Poster Sessions
Short Courses (full day) | Short Courses (half day) | Tutorials | Practical Computing Demonstrations | Closing General Session with Refreshments
Viewing Practical Computing Demos only — View Full Program |
||
Saturday, February 21 | ||
PCD1 Interactive Predictive Modeling with JMP 12 Pro: Keeping It in the Flow
Fill out evaluation |
Sat, Feb 21, 2:00 PM - 4:00 PM
Napoleon C3 |
|
Instructor(s): Mia L. Stephens, JMP Division of SAS; Scott Lee Wise, SAS Institute, JMP Division | ||
Interactive predictive modeling in JMP Statistical Software from SAS is more than building models. It allows you to take advantage of interactive and dynamic graphs and advanced analytic tools, keeping data visualization, analysis, and modeling in the flow. In this talk, we will use case studies to see how to explore and prepare data using the Column Switcher, Data Filter, Recode, and Graph Builder. We will use the Partition platform, Fit Model, and Generalized Regression platforms, as well as tools such as the Prediction Profiler and the Solution Path in JMP Pro 12, to interactively explore parameters and select potential models. Finally, we’ll compare a variety of competing models using Model Comparison.
|
||
PCD2 Tessera: Open Source Tools for Big Data Analysis in R
Fill out evaluation |
Sat, Feb 21, 2:00 PM - 4:00 PM
Borgne |
|
Instructor(s): Landon Sego, Pacific Northwest National Laboratory; Amanda White, Pacific Northwest National Laboratory | ||
Tessera is a set of R-based tools to enable data scientists to explore and analyze large, complex data.The Tessera computational environment is powered by divide and recombine (D&R), an approach for dividing data into subsets and computing on them in parallel. At the front end, the analyst programs in R. At the back end is a distributed parallel computa- tion environment such as Hadoop. In between are three Tessera packages: DataDR,Trelliscope, and RHIPE.The DataDR R package provides a high-level interface to D&R operations, making specification of divisions, analytic methods, and recombinations easy.The interface is designed to be back end agnostic, so it can harness new distributed computing tech- nologies as needed.Trelliscope is a scalable visualization tool in which data sets are divided into subsets and a visualization method is applied to each subset and shown in a multi-panel trellis display.This framework has proven to be a powerful mechanism for all data, large and small. RHIPE is the R and Hadoop Integrated Programming Environment. RHIPE allows an analyst to run Hadoop MapReduce jobs from within R. RHIPE is used by DataDR when the back end is Hadoop.
|
||
PCD3 Mathematica and Statistical Computing
Fill out evaluation |
Sat, Feb 21, 2:00 PM - 4:00 PM
Napoleon D1 |
|
Instructor(s): Michael Kelly, Wolfram Research Inc. | ||
Mathematica is the world’s leading symbolic and numerical software, pioneering the use of symbolic functional programming for the representation of mathematical, statistical, and computational objects in a universal, consistent, and high-level language that has allowed for a systematic treatment of the entire area of statistical analysis. Unlike other statistical programs that are mainly numerical, Mathematica combines the many advantages of symbolic representation of mathematical statistics with the numerical capabilities of advanced and novel algorithms. See www.wolfram.com/solutions/industry/statistics.
|
||
PCD4 Rating College Football Teams: A Case Study on Integrating Minitab with Statistical Programming Languages
Fill out evaluation |
Sat, Feb 21, 2:00 PM - 4:00 PM
Napoleon D2 |
|
Instructor(s): Daniel Griffith, Minitab; Eduardo Santiago, Minitab, Inc. | ||
College football is a sport with highly variable outcomes and teams that play highly unbalanced schedules due to conference affiliation, a large pool of potential opponents, and incentives that disfavor competitive balance. Despite these difficulties, it is highly desirable for fans, media, and the playoff selection committee to rate teams as accurately as possible. Using an unconventional method, the case study demonstrates how teams can be rated with minimal effect from uncontrollable aspects of the game.The method is performed using a combination of Minitab Statistical Software for its ease of use and graphical capabilities integrated with a statistical programming language for complex routines.
|
||