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



A Practitioner’s View of Predictive Modeling

Methods, Tactics, Tools and Techniques

Classroom: Two Days
Online: Four Half-Day Live Sessions
$1,795 USD
INFORMS Professional Development Units: 18


TMA’s Modeling Practice Framework™

The “Model Development” course dives into the data mining process at the tactical level. Participants will observe live demonstrations of machine learning methods and computer-aided pattern discovery techniques for extracting and interpreting complex patterns and relationships from large volumes of data. Attendees then participate in work-along labs that build upon an overall project.

This course is designed to be taken independently, yet is part of a larger course series that covers The Modeling Agency’s Modeling Practice Framework™ for low-risk, high-impact projects. The scope of this course extends to the fourth and fifth phases: Model and Validate.

You need not be an experienced statistician or mathematician to track well in this course – though traditional quantitative experts will benefit from the strategic referencing to the Plan phase and the pragmatic mind-shift required in this event. The machine learning algorithms are covered from a functional perspective. Modern software does a great job of handling the mathematical complexity. Those seeking a deep drill-down into the mathematical or theoretical underpinnings of predictive analytics algorithms should refer to other available academic offerings.


      • Data Scientists: who desire to extend their analytical toolbox and underscore the scientist aspect of the role with formal process and hands-on methodological practice
      • Functional Analytic Practitioners: Customer Relationship Managers, Risk Analysts, Business Forecasters, Statistical Analysts, Social Media and Web Data Analysts, Fraud Detection Analysts, Audit Selection Managers, Direct Marketing Analysts, Medical Diagnostic Analysts, Market Timers
      • Big Data Analysts: who are under increasing pressure to transform their deluge of data from a liability to an asset
      • Project Leaders: who desire to have a more detailed understanding of predictive modeling methods and techniques to better manage and interact with their practitioners
      • Business Analysts: who must develop and interpret the models, communicate the results and make actionable recommendations
      • IT Professionals: who wish to gain a better understanding of the data preparation, analytics and analytic sandbox development requirements to more fully support the growing demand for analytic IT support
      • Anyone Overwhelmed with Data and Starved for Actionable Insights


      • Apply a formal process for data preparation, model development and validation of results
      • Recognize and avoid common but costly pitfalls in data prep, method selection and results interpretation
      • Ensure that your model is adequately generalized and has not memorized the training set
      • Take an incremental low-risk / high-impact approach to model development along with vendor-neutral tool exposure that will save months in product surveying
      • Consider trade-offs between model accuracy and explainability when selecting modeling methods
      • Guide your IT staff to build an analytics sandbox for rapid model development and minimal IT dependency
      • Proceed confidently upon return with the formal process, session files, and direct hands-on experience gained through follow-along labs in the Model and Validate phases of the Modeling Practice Framework™.


The rapid emergence of data processing and collection methods has propelled the IT industry into the age of big data. Organizations now contain so much data that it has become very difficult to understand just what all that data is telling us.

There is hardly a transaction that does not generate a computer record somewhere. All this data has meaning with respect to making better prospective organizational decisions and anticipating customer needs and preferences. But how do you discover those needs and preferences within data stores that contain endless amounts of seemingly incomprehensible numbers and facts? Predictive analytics does just that.

Only predictive analytics can transform the rapidly growing mass of inert data into actionable information assets. These assets can be leveraged for measurable gain through improved and automated decision-making. The practitioners who develop these transformative skills will be the leaders in their field.

The intent of this course is to present participants with a roadmap for model-building, how various methods and tools apply to different kinds of data intensive problems, and how to overcome limitations that cause the majority of predictive models to under-perform.


    • Understand the purpose, function and impact of the Modeling Practice Framework™
    • Proceed through the general implementation of the two tactical phases: Prepare and Build
    • Realize that model-building for actionable production need not be highly technical or complex
    • Construct a valid data set and transform data for superior model performance
    • Describe each of the five steps for preparing raw data for predictive analysis
    • Select appropriate methods for each of the Four Core Analytic Project Types
    • Assess the degree to which a model meets a predefined performance objective
    • Leave with resources, contacts and actionable plans to substantially increase your analytic capabilities while minimizing dead ends


This course does not restrict or skew the presentation of data mining methods through a single product. Rather, the Model Development course gives broad consideration of the capabilities and limitations of all resources from a vendor-neutral perspective.

Live modeling demonstrations projected from the presenter’s machine will precede the follow-along lab exercises. The exercises run from a design produced in the Plan phase. Participants will directly experience the natural messiness of data mining to discover what really works… as well as what doesn’t and why. The instructor will show how to evaluate various features and available products based upon strengths, limitations, function, value, and general performance.

TMA’s highly seasoned faculty possesses a wealth of pragmatic experience in applying predictive analytics across industries in current real-world applications. This course, like no other, insists upon making predictive analytics purposeful, measurable and actionable in a business or organizational setting.


While this course is designed to be taken independently, it is important to understand its place and function within the overall Predictive Analytics & Data Mining Course Series.

Registrants will be required to view a four-hour asynchronous “Core Concepts” orientation prior to attending this event. Access details for the Core Concepts orientation will be shared with participants prior to the start of the course. Prior education or experience in data analytics or statistics is helpful, but not required.

Participants need only supply a laptop computer with Microsoft Excel. Instructions on how to download lab data and any analytic tools will be provided in the preparatory email. The instructor can assist participants with any preparation during breaks, and before or after class.


What You Will Get in this Presentation

Core Concepts
Prepare Phase
Model Phase
      • Build & Test Alternative Models
      • Evaluate Model Results
      • Model Accuracy and Stability
      • Lift and Gains charts
      • Modeling Demonstration
        • Assess Model Viability
      • Why Accuracy and Stability are Not Enough
      • What to Look for in Model Performance
      • Exercise Breakout Session
        • Select Final Models
        • Create & Document Modeling Plan
        • Determine Readiness for Deployment
      • Assessing Data Prep for Deployment
      • What are Potential Deployment Challenges for Each Candidate Model?
      • Exercise Breakout Session
Validate Phase
    • Select the Most Strategic Model Option(s)
    • Validate Finalist Models
    • Prepare Data for Test Deployment
    • Data Preparation Steps for Production
    • Data Preparation Demonstration
      • Measure Lift / ROI / Impact
    • The Potential Challenges of Estimating ROI
    • Designing an Effective “Dress Rehearsal”
    • The Basics of A/B testing
    • Exercise Breakout Session
      • Test Deployment
      • Document Validation Process
Wrap-up and Next Steps
          • Apply a 20% Series Discount within six months to the Project Planning course
          • Supplementary Materials and Resources
          • Conferences and Communities
          • Get started on a Project!
          • Strategic Oversight and Collaborative Development

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.


Upcoming Sessions

November 2 & 3, 2016
December 5 – 8, 2016
12:00p – 5p US EST | 9:00a – 2p US PST
17:00 – 22:00 UTC / GMT
February 13 – 16, 2017
12:00p – 5p US EST | 9:00a – 2p US PST
17:00 – 22:00 UTC / GMT

Sign Up Early and Save

Events Limited to 20 Seats

On-Site Available

Inquire by Phone or Email

Predictive Analytics Webinar

Learn How to Get
Predictive Modeling Off
the Ground and Into Orbit

Next Event

Tuesday, November 8, 2016
4pm EST / 1pm PST
90-Minute Live Interactive Event

Why Train With TMA?

Determine whether TMA training is right for you, and learn why TMA is truly the best option for live classroom analytics training.

Attendee Comments

“The instructor is knowledgeable, well organized, and interacts extremely well with participants. If you have only two days to learn about data mining, TMA’s Model Development course is the class you should attend.”

Yiguang Qiu, PhD
Marketing Department
Amica Insurance

“This course is a great overview of available predictive analytics methods and techniques that provides an organizing framework for your model development efforts.”

April Smith
Lead Workforce Analytics

“Both instructors in the series did a fantastic job of getting me up to speed in predictive analytics much faster than any book (or probably any other training class or conference) available.”

Raymond G. Henderson
Knowledge-Based Systems Engineer
Compliance Technologies, Inc.

“The scope and pace of the class were spot on for me, exactly what I was looking for. The real world examples from industry veterans were both insightful and beneficial. The class ultimately helped to strip away the buzz words and hype surrounding the industry, and revealed the truly valuable foundation of this remarkable combination of art and science.”

Eric Vajentic
Senior Business Analyst
Cerner Corporation

“This is a must-attend course for those who would like to get started in the amazing world of Predictive Analytics & Data Mining. Without going into theoretical details, it covers every important step in the development of predictive models.”

Ivan Mendoza
Database Marketing & Predictive Analytics Expert
Cogeco Cable

“This course was very enlightening. I was suprised to see the many ways predictive analytics can be applied!”

Roxanne Ramos
Business Operations Analyst
Toshiba America Medical Systems

“The wealth of information covered in these courses, as well as the in-depth demonstrations of multiple software packages, made the sessions valuable from a wide range of perspectives. I will certainly recommend that others attend.”

Brent King
AVP, Managed Care
Analytics / Business Development
Health Smart Preferred Care

“The instructor was great and the instructor was very knowledgeable. It was just at the right level, not to general and not too technical.”

Doug Rosenberg
Technical Consultant

“The instructor and course material are first rate. Any organization that believes data mining should be a part of their business operations portfolio would be making a wise investment by attending this course.”

Eric Rickard
Information Computing Sciences
SRI International

“The instructor was really great. Tim not only had a very clear understanding of what he was presenting, he was also a very good teacher who solicited a lot of audience participation. Most people with a highly technical background can’t always communicate effectively. Tim had no such problem.”

Bryan Arnold
Current Employment Statistics
Bureau of Labor Statistics

“The Modeling Agency takes a daunting subject and brings it to a very understandable level to attack any problem you may be facing as an organization.”

Eric Barger
Personal Lines Underwriting Manager
Brethren Mutual Insurance Company