This course teaches the concepts required to effectively utilize the broad array of computational features available in the TIBCO Spotfire Analyst client.

Learn about the computational capabilities which can be executed from within the TIBCO Spotfire Analyst environment. These include calculated columns, custom expressions, curve fitting, curve drawing, executing statistical code on TIBCO Enterprise Runtime for R (TERR), or external engines such as S+, R, SAS, or MATLAB. In addition, you will learn how to discover and evaluate data relationships, as well as generate predictive models or apply computational multivariate data analysis tools such as similarity calculations or clustering techniques.

Audience: Power users who need to use customized expressions, curve fitting, predictive modeling, R statistical code, data functions, or multivariate data analysis tools in order to extend analysis capabilities
Course Duration: 4 days
Prerequisites:

We recommend students participate in Spotfire Analyst Essentials I and Spotfire Analyst Essentials II prior to taking this course.

Course Outline:
  • Spotfire Expressions
    • Spotfire Expressions Overview
    • Introduction to Spotfire Expressions
    • Including OVER in Calculated Columns
    • Including OVER in Custom Expressions
    • Collect Property Values for Expressions
    • Use Property Controls to Select Custom Expressions
    • Functions for Use in Expressions
    • Understand Expression Shortcuts
    • Performing Calculations Across Data Tables
    • Defining Categorical Variables and Hierarchy Levels in Expressions
    • Build Expressions and Understand Expression Syntax

 

  • Statistical Engines
    • Statistical Engines Overview
    • Execute TERR Functions within Spotfire Expressions
    • Remove Replicates Using a TERR Expression
    • Register and Run Data Functions
    • Data Function Properties – Online vs Offline
    • Using Sample Data Functions
    • Data Functions Review

 

  • Relationships and Predictions
    • Relationships and Predictions Overview
    • Apply Horizontal or Vertical Lines to Visualizations
    • Apply Curve Fit Algorithms to Visualizations
    • Calculate IC50 Values Using Logistic Regression
    • Options to Draw Lines on Visualizations
    • Perform LOWESS smoothing with TERR
    • Draw regression line confidence intervals with TERR
    • Forecast Values Using the Holt-Winters Method
    • Discover Relationships Between Numerical vs. Numerical Columns
    • Discover Relationships Between Numerical vs. Categorical Columns
    • Discover Relationships Between Categorical vs. Categorical Columns
    • Use Regression Modeling to Predict Numerical Values
    • Use Classification Modeling to Predict Categorical Assignments
    • Save and Share Predictive Models Across Analyses

 

  • Multivariate Data Analysis
    • Multivariate Data Analysis Overview
    • Visualizing Multivariate Data
    • Apply Line Similarity to Isolate Patterns of Interest
    • Apply K-means Clustering to Group Similar Lines
    • Cluster Stock Data Collected Over Time
    • Apply Hierarchical Clustering to Organize Multivariate Data
    • Interpreting and Interacting with Dendrograms
    • A Quick Review of Hierarchical Clustering
    • Performing Principal Component Analysis Using TERR
    • Normalizing Data in Preparation for Computational Multivariate Data Analysis
    • Consider Empty Values when Applying Computational Multivariate Data Analysis