Regression Analysis Executive Overview (4 hours)

Description

Regression analysis provides a quantitative relationship between two or more variables that can be used for making predictions. This course covers the basic elements of performing and interpreting regressions. Students will learn how to choose the right sort of regression for different data sets, how to derive a regression from raw data, how to check to see if the regression accurately reflects the relationship in the data, and how to make statistically relevant predictions from the regression. Managers and employees responsible for the interpretation of information used in process improvement efforts will benefit most from this course.

The course length is 4 hours and will include detailed instruction, periodic assessments, and a certificate of completion upon successful review.

Lesson 1: Linear Regression

• Define the goal of linear regression
• Define the linear regression formula
• Interpret a scatter diagram
• Use the correlation coefficient, the coefficient of determination, the standard error of estimate and the significance of slope to determine how well a regression fits the data
• Plot and examine residuals, and make decisions about the validity of the model based upon them
• Calculate and interpret the confidence interval for the individual and the mean value

Lesson 2: Polynomial Regression

• Describe the factors to consider when deciding whether to use linear or polynomial regression
• Use the polynomial regression model to predict a dependent variable
• Test the significance of the polynomial regression model
• Interpret the correlation coefficient, the coefficient of determination, the significance of slope and the standard error of estimate for a polynomial regression
• Check residuals to judge the accuracy of your model
• Use confidence intervals around a polynomial regression model to estimate the precision of an estimated value

Lesson 3: Multiple Regression

• Identify when the use of multiple regression is appropriate
• Use the multiple regression model to predict a dependent variable
• Test the significance of the multiple regression model
• Use the correlation coefficient, the coefficient of determination, the significance of slope and the standard error of estimate for a multiple regression
• Check residuals to judge the accuracy of your model
• Use confidence intervals around a multiple regression model

Target Audience

• Operations Manager
• General Manager
• Plant Manager
• Person with bottom-line responsibility

UL University's workshops are designed for and intended to serve individuals using and relying upon UL services. UL University reserves the right to limit workshop attendance strictly to the foregoing. In addition, UL University reserves the right to change, reschedule or cancel any workshop at any time. UL shall not be responsible for any consequential or other losses resulting from the cancellation or postponement of this workshop such as airline and other travel/personal expenses.

UL University accepts the following forms of payment for registration: Visa, MasterCard, American Express and Invoice/PO. Please note that Invoice/PO is not accepted for online courses.