Businesses have access to more data than ever before. Predictive analytics is the process of using historical data to create actionable insights about future events or outcomes. There are several approaches to predictive analytics, including classification modeling, machine learning, data mining and regression analysis. This short course will provide an overview of various predictive analytics techniques and a hands-on introduction to regression analysis using Excel. Participants will leave this course with the tools needed to help guide data preparation and analysis to facilitate quantitative predictions for their organization.

By the end of this hands-on course, you will be able to:

• Sort and filter data
• Create "indicator" variables
• Create scatter plots and time series plots
• Calculate summary statistics to describe data
• Calculate and interpret correlation coefficients
• Calculate and interpret the line of best fit
• Forecast/predict future values

This program is designed for people who:

• need to quickly producing reports in Excel 
• have basic exposure to Excel
• need to analyze data in Excel to improve decision-making 

Goals for prediction

• Defining the project and selecting the dependent variable
• Forecasting vs. interpolation

Data collection and dataset preparation

• Sampling
• Types of data
• Data transformation and preparation
• Outliers and missing values

Model development and model selection

• The ordinary least squares regression model
• Selecting independent variables
• Understanding goodness of fit
• Interpreting the impact of independent variables

Using the model to make prediction and inference

• Forecasting and calculating fitted values
• Margins of error – specifying confidence in your predictions
• Cautions, caveats and concerns

coming soon

Peter Schuhmann, Ph.D.
Dr. Peter Schuhmann teaches economics at the University of North Carolina Wilmington. He earned his Ph.D. in economics from North Carolina State University in 1996 with field concentrations in environmental economics and statistics.

Dr. Schuhmann has taught six unique undergraduate courses, three graduate-level courses and has supervised over 100 graduate and undergraduate independent studies and theses. He teaches traditional face-to-face courses, hybrid courses, and fully online courses and was the co-organizer of an annual regional economics teaching workshop held in Wrightsville Beach, North Carolina for 15 years.

Dr. Schuhmann's primary area of research is the non-market valuation of environmental goods and services, largely focused on coastal and marine resources in North Carolina and the Caribbean. His research includes analysis of willingness to pay for changes in coastal and marine resource quality, including beach width, beach amenities, reef health and species diversity, as well as examinations of the costs and benefits of fisheries policy and shoreline management. Dr. Schuhmann's work has been published in journals such as Ecological Economics, Land Economics, Marine Resource Economics, Marine Policy, Natural Resource Modeling, and the Journal of Environmental Management.