Tuesday, 8:30 AM
|
2019-US-TUT-290
Essential Scripting for Efficiency and Reproducibility: Do Less to Do More
Drew Foglia
Evan McCorkle
From reproducing simple tasks to automating daily processes to sharing scripts with colleagues to deploying full applications across your organization, challenges exist at every level that can limit the efficiency and reliability of your JSL scripts. In this tutorial, we will travel along the arc from small scripts to large applications presenting best-practice techniques for mitigating many of the common, yet subtle, pitfalls that often hinder JSL novices and veterans alike. We will discuss strategies for combining multiple steps into a cohesive script, and conversely splitting overly large scripts into more manageable files. We will also present tips for handling errors, wrangling windows, isolating variables from unexpected changes, protecting the integrity of your scripts without compromising usefulness and more.
|
2019-US-TUT-291
Advanced Data Preparation: 10 Essential Tools in JMP® to Get From “Messy” to “Analysis Ready”
Julian Parris
Rarely, if ever, do data come to us in an “analysis ready” format. Luckily, JMP has a rich and expansive set of tools that enable you to efficiently prepare your data for analysis. In this tutorial we explore 10 of the essential tools in JMP that help us get our data from “messy” to “analysis ready,” including methods for handling table restructuring and joining, computed and derived variables, outliers and influential points, recoding of variables, missing values and more. After we explore each of the 10 essential tools in depth and discuss best practices (and even some “off-label” uses for certain tools), we’ll work through three case studies where we will apply these tools in various ways to efficiently import, recode, restructure and reorganize complex and challenging data sets. Previous experience using JMP is highly recommended, though not strictly necessary.
|
2019-US-TUT-270
Which Model When?
Ruth Hummel
Mary Loveless
You have a business or research question, you’ve collected or found appropriate data, and you are ready to analyze. But which analytical methods should you try? And how will you choose a final model? In this talk, we will look at several data scenarios and present modeling options and a framework for comparison. We will look at how different questions or goals affect the modeling choices we make. (Predict? Explain? Find associations?) Models covered will include traditional regression, penalized regression, partial least squares and a few others. Comparison techniques will include residuals analysis, comparing fit statistics, and cross-validation or validation on new data.
|
Tuesday, 10:30 AM
|
2019-US-TUT-264
Before the Modeling: Feature Engineering in JMP(R)
Jordan Hiller
Mike Muhlada
Successful predictive modeling projects require careful feature engineering. Both a science and an art, feature engineering involves gathering, summarizing, reshaping and transforming data into a form usable by modeling algorithms. This tutorial will describe some of the common activities in feature engineering and demonstrate JMP tools that can be used to perform them. There will be in-depth discussion and examples showing the use of formula columns for binning and transformation, the Recode platform for collapsing categories, and Query Builder for summarization.
|
2019-US-TUT-292
Graph Builder: From Discovery to Presentation
Bill Worley
Graph Builder is a jack-of-all-trades when it comes to visualizing your data. It is useful throughout the analysis process, from exploring your data to understand what you've got, to discovering relationships, to crafting the perfect graph to present your results. This session will show you how to use Graph Builder effectively with an overview of the graph elements available and the options for each one. It will also show you some tricks to help you become a master graph builder.
|
2019-US-TUT-293
The Most Flexible Modeling Platform That You're Not Using
Clay Barker
Generalized Regression — it sounds scary, but it's not. This platform in JMP Pro can handle most common linear modeling exercises. Need to find the best set of predictors in a sea of possibilities (i.e., variable selection)? This platform can do that. Need to model a categorical response with more than two levels? Generalized Regression can do that. Have problems with non-normal responses like count data or yield percentage? Where other modeling methods fall down with these responses, Generalized Regression handles them with aplomb. If you're not using this platform, you're missing out. We will look at using Genreg to build models in a variety of settings: starting with an orthogonal designed experiment and moving to large observational data sets.
|
Tuesday, 1:00 PM
|
2019-US-TUT-246
Building Dashboards in JMP(R)
Dan Schikore
JMP platforms are often used in combination with one another to help guide a workflow or to provide complementary analysis and visualization of the data. The JMP Dashboard Builder can help you organize multiple reports in a single window and reproduce the set of reports using the same data table or a new table. Dashboard layout can be done automatically or through interactive drag-and-drop operations, and one report can optionally be used to filter others in the dashboard. Dashboards can be saved to a JMP data table to reproduce the reports with the same table, or they can be saved to an add-in to share the dashboard with colleagues and reproduce the same report on different data tables. When dashboards depend on the results of database queries, you have the choice to re-run the query each time, or to use a saved copy of the query. Dashboard results can also be saved to JMP Public or a private JMP Live server to share interactive reports with others via the web.
|
2019-US-TUT-287
The Fundamentals of Modern Experimentation Using JMP®
Bradley Jones
Designed experimentation is the best way to learn about industrial processes. But, in the long history of the design of experiments, most experimenters learned methods that focused on conforming the experiment to classical designs. The need to understand these constraints limited the number of effective experimenters. JMP’s approach to DOE takes the burden of understanding the constraints of textbook designs off the practitioner, instead putting the experimenter in control. See how to use the Custom Designer to design an experiment for any process. The Custom Designer creates an experiment tailor-made for your situation. Do you have constraints on the input factors? No problem. Mixture factors along with process inputs? No problem. What about a run budget that limits the number of experiments you can perform? The Custom Designer will give you a design that will allow you to learn as much as possible within your budget. A few years ago Stu Hunter said that, "[the technology behind the DOE platforms in JMP] would change the way I teach DOE.” Come see how.
|
2019-US-TUT-193
Large-Scale Process Monitoring Using JMP(R)
Laura Lancaster
Jianfeng Ding
In this age of big data and complex manufacturing there is often an enormous amount of process data that needs to be monitored and analyzed to maintain or improve quality. JMP has several tools to help the analyst quickly and efficiently increase the scale of their process monitoring. The Process Screening platform allows users to easily scan processes for stability and capability and enables them to focus attention on processes needing improvement. The platform initially computes a summary report based on control chart, capability, and stability calculations and creates several graphs for quick visual assessment of process health. Based on these initial results, it is easy to select the processes needing attention and explore them more in depth with access to Control Chart Builder and Process Capability. The Model Driven Multivariate Control Chart (MDMCC) platform, new in JMP 15, allows users to monitor large amounts of highly correlated processes. This platform can be used in conjunction with the PCA and PLS platforms to monitor multivariate process variation over time, give advanced warnings of process shifts, and suggest probable causes of process changes. We will use case studies to demonstrate how to use JMP to monitor and analyze many processes for fast and efficient improvement.
|
Tuesday, 3:00 PM
|
2019-US-TUT-294
Getting the JMP(R) on Your Clinical Trial Analysis
Geoff Mann
Chris Kirchberg
The latest JMP Clinical version 7.0 was released in fall of 2018. JMP Clinical is a vertical solution leveraging JMP capabilities and an extensive JSL infrastructure to create a customized interface for clinical trial data reviews for medical safety, early efficacy, data integrity and oncology analysis. JMP Clinical 7.0 represents significant enhancements that take advantage now more than ever of key features in JMP 14. In this presentation, we will present an overview of the new features of JMP Clinical, focusing heavily on a new comprehensive review subject filter built around virtually joined JMP tables using row state synchronization (new in JMP 14) and data filters. Our JMP Clinical reviews provide standardized, reproducible analyses and now we also introduce an infrastructure for saving further JMP customizations into our reports to allow you to tie in your own JMP analyses to our applications. We also highlight new features in JMP Graph Builder to provide highly customized visualizations for solid tumor oncology clinical trial analysis, including swimmer plots, waterfall plots and spider plots. More generally, we will discuss how this vertical solution exemplifies building robust applications via JSL and the JMP add-in architecture to enhance the use of JMP.
|
2019-US-TUT-295
Let's Talk Tables
Kelci Miclaus
Mandy Chambers
JMP 13 introduced virtual joins for JMP tables. This capability enables you to use common keys to link multiple tables, without using the time/memory necessary to actually create a joined (denormalized) copy of your data. Virtually joining tables gives a table access to columns from the linked tables for easy data exploration. A new feature in JMP 14 advances the virtual join capabilities to allow linked tables to communicate with each other by sharing rows states. New options allow you to set up a link reference table to listen and/or dispatch row state changes among virtually joined tables. This feature provides an incredibly powerful data exploration interface that avoids unnecessary table manipulations. This presentation will highlight the new features in JMP with examples using the common sample data followed by a practical application of these features as implemented in JMP Clinical. We will create a review of multiple metrics on patients in a clinical trial that are virtually linked to a subject demographic table and show how a data filter on the Link ID table enables global subject filtering throughout all of the linked safety (adverse events, labs, etc.) tables.
|
2019-US-TUT-289
Introduction to Functional Data Analysis
Chris Gotwalt
Ryan Parker
JMP Pro is now able to model and visualize functional data in a new way that is direct, straightforward and leads to highly accurate models. We will introduce the Functional Data Explorer platform and go through a set of examples that illustrate two particularly useful families of problems that it facilitates solving, function response DOE analysis (FR-DOE) and functional regressor machine learning (FR-ML). FR-DOE analysis allows one to see directly how the shape of a response curve changes as a result in changes in the value of the DOE factors. FR-ML is an approach to feature extraction for sensor data from manufacturing processes that faciliates accurate early prediction of final batch yield or faulty products. Our examples will be primarily from pharmaceutical, chemical and semiconductor manufacturing, but our approach can be applied to practically any industry.
|
