Menu

Learning Track

Predictive Analytics

This 3-day track provides participants with a comprehensive toolkit to effectively apply predictive analytics in their organization.

It is appropriate for members of a problem-solving team, those leading operational excellence activities, marketing analysts, and practitioners preparing to implement predictive analytics in their organization.

Learn the data analysis techniques necessary to analyze patterns found in historical data to gain better insights, identify potential risks, seek out improvement opportunities, and make predictions about the future. Analytics principles will be presented through real-world examples and exercises – all supported by Minitab Statistical Software.

Schedule On-Site Training
Find Public Sessions

DAY 1

Fundamentals of Analytics

In this foundational course, you will learn to minimize the time required for data analysis by using Minitab to import data, develop sound statistical approaches to exploring data, create and interpret compelling visualizations, and export results. Automate your Minitab analysis with minimal user input to save time. Analyze a variety of real-world data sets to learn how to align your applications with the right analytics tool and interpret the statistical output. Learn the fundamentals of important statistical concepts such as hypothesis testing and confidence intervals.

This course places a strong emphasis on making sound decisions based upon the practical application of statistical techniques commonly used in business, manufacturing, and transactional processes.

Topics Include:
 
  • Importing and Formatting Data
  • Exec Macros
  • Bar Charts
  • Histograms
  • Boxplots
  • Pareto Charts
  • Scatterplots
  • Measures of Location and Variation
  • t-Tests
  • Test for Equal Variance
  • Power and Sample Size
Participants of the course will be able to:
 
  • Easily organize and visualize data
  • Automate analysis by writing and running Exec Macros
  • Make decisions about process parameters using statistical methods

Pre-requisites: None

Audience: Data Scientists, Analysts, Engineers, Scientists, Business Professionals, Operational Excellence Professionals

DAY 2

Regression Modeling and Forecasting

Continue to build on the fundamental statistical analysis concepts taught in the Fundamentals of Analytics course by learning to explore and describe relationships between variables with statistical modeling tools. Discover and describe features in data related to the effect and impact of time, and how to forecast future behavior.

Learn how to find and quantify the effect that input variables have on the probability of a critical event occurring. Hands-on examples illuminate how modeling tools help reveal key inputs and sources of variation in your data.

Topics Include:
 
  • Scatterplots
  • Correlation
  • Simple Linear Regression
  • Time Series Tools, including Exponential Smoothing
  • Trend Analysis
  • Decomposition
  • Multiple and Stepwise Regression
  • Binary Logistic Regression
  • Regression with Validation
Participants of the course will be able to:
 
  • Use correlation and regression tools to model variable relationships
  • Forecast using a time series model
  • Model the relationship between predictors and a binary response

Pre-requisites: Fundamentals of Analytics

Audience: Data Scientists, Analysts, Engineers, Scientists, Business Professionals, Operational Excellence Professionals

DAY 3

Machine Learning

Expand your analytics by analyzing data from real world problems experienced in many industries to explore and describe relationships between variables. Learn to use supervised machine learning techniques such as CART® to analyze patterns found in historical data to gain better insights, identify potential risks, seek out improvement opportunities, and make predictions about the future.

Use unsupervised machine learning tools such as Clustering to detect natural partitions in the data and group observations or variables into homogenous sets. Reduce the dimensionality of data by transforming the original data into a set of uncorrelated variables.

Topics Include:
 
  • Discriminant Analysis
  • Test Set Validation
  • K-fold Validation
  • CART® Classification
  • Correlation
  • CART® Regression
  • Cluster Analysis
Participants of the course will be able to:
 
  • Learn how to apply unsupervised learning tools to segment the data
  • Use CART Classification to create a decision tree to model the relationship between a categorical response variable and predictors variables
  • Model the relationship between a continuous response variable and many predictor variables using CART
  • Regression to reveal important patterns within highly complicated data

Pre-requisites: Fundamentals of Analytics, Regression Modeling and Forecasting

Audience: Data Scientists, Analysts, Engineers, Scientists, Business Professionals, Operational Excellence Professionals

clear