• This field is for validation purposes and should be left unchanged.
Guidance and results for the data-rich, yet information-poor.

SERIOUS PLAY FOR PREDICTIVE ANALYTICS

What Works, What Doesn’t and Why

One Day Seminar
$795 USD
NASBA CPEs: 6


ABOUT THIS SEMINAR

This one-day vendor-neutral event will prepare analytic practitioners and functional managers to make sense of predictive modeling and take control of the analytic process. “Serious Play for Predictive Analytics” introduces the foundation for data-intensive analytic projects that deliver insight, clarity, confidence and actionable decision support.

Live demonstrations will illustrate how organizational practitioners can effectively maneuver the natural messiness of advanced analytics. Attendees will realize that true impact with predictive analytics has far more to do with the overall management of a project team and strategic process than with the tactical skills of a data scientist.

If you are a business or public sector practitioner or leader seeking to propel your organization’s analytic maturity and put predictive analytics to work for measurable gain, then this event is designed for you.


WHO SHOULD ATTEND

    • IT EXECUTIVES AND BIG DATA DIRECTORS: CIOs, CAOs, CTOs, Functional Officers, Technical Directors and Project Managers who desire to transform their deluge of inert data to actionable assets
    • LINE-OF-BUSINESS DIRECTORS AND FUNCTIONAL MANAGERS: Risk Managers, Customer Relationship Managers, Business Forecasters, Inventory Flow Analysts, Financial Forecasters, Direct Marketing Analysts, Healthcare and Bioinformatics Analysts
    • DATA SCIENTISTS: Who recognize the importance of complementing their tactical proficiency with a strategic planning and design approach to advanced analytics
    • TECHNOLOGY PLANNERS: Who survey emerging technologies in order to prioritize corporate investment
    • CONSULTANTS: Whose competitive environment is intensifying and whose success requires competency with predictive analytics, big data and related emerging technologies

THE BUSINESS CHALLENGE

Traditionally, organizations use data retrospectively – to view what has already happened. Leading organizations use data prospectively – to anticipate behavior and automate prescriptive decisioning that targets the allocation of valuable resources while minimizing risk and loss. The mining of data for predictive indicators creates information assets from big data or small, which an organization can leverage to achieve specific strategic objectives.

Predictive analytics may be defined as “the data-driven discovery and modeling of hidden patterns in large volumes of data.” The resulting models are both descriptive and prospective. They address why things happened and what is likely to happen next. The organizations that effectively transform their big data liability into information assets and automate decision-making for measurable gains will be the first to realize substantial returns on their big data and analytic investments.


KEY SKILLS YOU’LL TAKE AWAY IMMEDIATELY

    • Develop a business-aligned strategy for applying high-value data-driven decisions
    • Identify, qualify and prioritize viable and actionable analytic opportunities
    • Convey a standardized process development model to implement across your team
    • Acquire both tactical and strategic skills required to stand out in the analytics practice
    • Learn why most analytics projects fail and the main pitfalls to avoid
    • View a standardized process methodology for predictive analytics
    • Leave with resources, contacts and plans to reduce your project preparation time, costs and risks

WHAT MAKES THIS SEMINAR UNIQUE

This aggressive one-day vendor-neutral event covers a host of critical considerations when approaching predictive analytics – or correcting a chronically flawed analytic practice. For today’s organizations to transform their growing mass of data into measurable and sustainable gains, they will have to first take a more methodical and holistic approach to predictive analytics.

This will require a purposeful and balanced approach to strategy and tactics. Most organizations jump directly into data and tools that tend to produce good models… then fail at the project level for a host of strategic reasons. Those who make the investment to fully assess their environment, situation, resources and objectives across all team members will produce project designs that result in analytic projects that are measurable, accountable, actionable and impactful.

TMA’s Modeling Practice Framework™

This seminar highlights TMA’s Modeling Practice FrameworkTM that applies equal emphasis to strategic and tactical issues. Leaders who take this comprehensive course will interact more effectively with their teams at the tactical level, while analytic practitioners will complement their existing algorithmic background with a more strategic focus.

This seminar is intended for those who insist upon establishing a fully effective modeling process on the first pass — or finally overcome chronic analytic failings.


TOPIC COVERAGE

What You Will Get in This Presentation

INTRODUCTION

    • Predictive Analytics Definition & Core Concepts
    • Statistics vs. Predictive Analytics: Complimentary Technologies
    • Goal-Driven Analytics
      • “The main thing is to keep the main thing the main thing.”
      • What is the goal of the analysis project?
      • What are the performance metrics for evaluating success of the decision process?
      • What is the behavior that impacts performance?
      • Is there sufficient data for the target behavior to develop an adequate model?
      • The Modeling Practice FrameworkTM
      • The Analytic Project Team
      • Analytic Opportunity Identification
      • The Advent of Data Science
          • The Arena: From Business Unit-Based to IT Department-Based
          • The Professionals: From Analyst to Data Scientist
          • The Analyses: From Descriptive Analyses/Business Intelligence to Predictive Analyses/Data Mining/Machine Learning
      • What is Predictive Analytic’s Role in Big Data?
          • Market Perceptions of Big Data
          • Big Data Needs Advanced Analytics…But Does Analytics Really Need Big Data?
        • What is Big Data’s Business Value?
            • Retail Use Case
            • Guerrilla Marketing Use Case
            • Medical or Government Use Case
            • ROI of Big Data and Associated Analytics
            • The Future of Big Data and Advanced Analytics

Phase 1: ASSESS

    • Comprehensive Project Assessment
        • Organizational Objectives
        • Motivation and Alignment of Leadership
        • Behavior(s) of Interest
        • Environmental Constraints
        • Operational Requirements
        • Identification of Scarce Resources
        • Threats to Project or Process
        • Defining Baselines and Evaluating Project Potential

Phase 2: PLAN

    • Project Definition: The Blueprints for Actionable Analytics
    • The Three Steps of Model Development
        • Train
            • Construct Candidate Models
            • Sample Size Requirements
            • Matching Modeling Methods to Project Type
        • Test
            • Decision Cycle Identification
            • Sample Size Requirements
            • Performance Evaluation of Candidate Models
        • Validate
            • Operational Decision Consistency
            • Strategy Specification
            • Validation Study Requirements

Phase 3: PREPARE

    • Know Your Data and How it Was Generated
        • Importance of Face-to-Face Interviews with those Close to Data Collection
        • Difficulty of Obtaining Appropriate Data
        • Data is Never Presented on a Silver Platter
        • What Data Should I Include in My Analytic Sandbox?
        • Some Data is Not Math-Compatible
        • What Does the Outcome or Target Variable Look Like?
        • What Data Representations Should I Use?
        • What Data Transformations Should Apply?
        • How Do I Select Variables for My Model?
            • Beware of Dependent Variables Masquerading as Input Variables
            • Example: Response to Credit Card Solicitation vs. Number of Plastics Used
        • How do I construct the Train / Test / Validate data sets?
        • Structuring Data for Modeling

Phase 4: MODEL

    • Process Objectives and Goals
    • Experimental Design: TRAIN Revisited
    • Selecting Condition Attributes
    • Analytic Model Assessment
        • Statistics
        • Tables
        • Graphs
        • Resampling / Bootstrapping
    • Ensemble Modeling Conceptualization
    • Bias – Variance Tradeoff
    • Classification Models
        • Logistic Regression
        • Decision Trees / Boosted Trees / Random Forests
        • K-Nearest Neighbor
        • Neural Networks
    • Forecasting Models
        • Linear Regression
        • Bayesian Regression
        • Neural Networks
    • Multiple Models are Usually Needed
    • Perfect Correlation is Not a Good Thing
    • and No Correlation is a Waste of Time

Phase 5: VALIDATE

    • Does Our Math Make Business Sense?
    • Organizational Performance is the Only Priority
    • Analytic Metrics Do Not Equal Organizational Performance Metrics
    • Establish a Model Competition
    • How to Pick a Challenger
    • Confirming That a Valid Challenger Was Selected

Phase 6: DEPLOY

    • Evaluating the Expected Performance of our Challenger
    • Adoption by Domain Experts
    • Adoption by the Operational Environment or End Users
    • Adoption by Leadership and Stakeholders
    • Project Failure is Not Our Worst Outcome…

Phase 7: MONITOR

    • Adapting to a Changing Environment
    • The Environment Always Changes
    • Our Organizational Goals Also Change
    • Measuring Primary Model Performance Degradation
    • Determine When to Install A Hot-Spare Challenger Model
    • Determine When to Refresh the Full 7-Phase Model Development Cycle

SPECIAL TOPICS

    • The Complexity of Large-Scale Analytics
        • Start with the Low-Hanging fruit: Structured Data
        • Unstructured Data May be 90% of Overall Content, But Usually Holds Only 10% of the Value
    • Specialization in Project Teams
    • The Power of Adapting Core Analysis Skills
    • The Even Greater Power of Honing Soft Skills
    • Where to Go From Here
    • Resources to Get You On Your Way

RESOURCES

    • Analytic Glossary
    • Recommended Books
    • Linkedin Groups
    • Data Repositories
    • Predictive Analytics Across Social Media
    • Webinars, Courses, Conferences

The Modeling Agency, LLC, is registered with INFORMS (the INstitute For Operations Research and the Management Sciences) as a Recognized Analytics Continuing Educational Provider for the CAP® (Certified Analytics Professional) program. The CAP® credential provides analytics professionals with a means to distinguish themselves and demonstrate to employers, colleagues, and the public that they are knowledgeable analytics professionals. Courses provided by The Modeling Agency, LLC, are automatically accepted by INFORMS when claimed by credential holders as evidence of continuing education. For more information about the CAP® program, including requirements, eligibility, benefits, preparation, and exam dates, please visit the INFORMS website at www.informs.org/certification.

seminar-image


Sign Up Early and Save

Events Limited to 20 Seats

On-Site Available

Inquire by Phone or Email

Attendee Comments

“This event gave me a new perspective on techniques and applications software that our federal agency had not previously seen. The content was great, and the very knowledgeable instructor kept the students attention by using real-life examples and discussion of additional resources. I highly recommend this course!”

Larry P. Taylor
AuditorUS Department of Education

“When the only complaint is that the seminar could be longer, I think you’ve got an excellent class! I very much enjoyed the instructor’s use of a real data set to demonstrate principles taught throughout the event. The instructor went out of his way both before and during the class to help me to translate the material to my own work.”

Susan Glass
Senior Engineer, Biological Technologies Analysis Solutions Wyeth

“The instructor is also a consultant. He fully understands real world scenarios, and a project approach to effective execution.”

Jean M. Cotis
Application Development ManagerUPS Freight

“I look forward to applying what we learned during this seminar to current and new projects. We will be more organized, efficient and effective by applying the concepts and techniques delivered in this class.”

Sean Schmitt
Analyst US Customs and Border Patrol

“No fluff at all. The instructor consistently provided specific, practical and understandable advice about complex ideas and processes. He explains things very clearly, backs up his answers with real-world examples and is patient with questions.”

-Anonymous

“As someone without a formal degree, analytics can be a daunting word. This event is approached in a very consumable way and without the blunt theory of academia. It is practical and applicable, which is exactly what analytics needs in order to support organizational objectives.”

Carmen Schwesinger
Analyst Government of Alberta