391-What Every Engineer Should Know About Regression Analyses
Having trouble opening course PDF documents?
Course Documents, Test Previews and other resources above may attempt to open or download Adobe PDF files. Most modern browsers have built-in PDF readers. If you have problems opening this file in your browser, check your browser's or computer system's settings. You can also Download a copy of Adobe PDF reader.
This continuing education course is written specifically for professional engineers with the objective of relating to and enhancing the practice of engineering.
Modern computing technologies and Big data have significantly changed the discourse on data mining and data efficacy. In any system where quantities change, it is of interest to look at the effects if any, of the system variables. Indeed, there may be a relationship (in our case statistical relationship) which may be approximated by a simple mathematical relationship. At other times, the mathematical or functional relationship may be complicated. Still there may be situations where there does not seem to be meaningful relationships between the variables and yet we might want to express or relate those variables by some sort of mathematical equations.
Regression Analysis is one of the most important statistical techniques used for data mining applications. It is a statistical methodology that helps estimate the strength and direction of the relationship between two or more variables, more specifically regressor and response variables and provides detailed insight that can be applied to further improve system outcomes. The importance of regression analysis lies in its singular focus on data which means numbers and figures that ultimately define a business entity. In Regression Analyses, two types of variables are of major concern, namely the regressor or predictive variables also known as independent variables, and the response variable.
The independent or predictor variable is one that is not random but is controlled (sometimes observed such as the amount of rainfall on a plot of land when the interest is on the effect of rainfall on crop yield) during an experiment. The dependent or response variable cannot be controlled but is rather measured as an outcome of the manipulation (or observation in the case of rainfall) of the independent variable and thus is a random variable. In this course, we will focus primarily on the following elements of Regression Analyses, namely:
- Parameters & Estimates
- Probability Distribution of the Parameters
- Covariance between two variables
- Simple hypothesis tests involving parameters including one- and two-sided t and F tests
- Confidence Interval for the parameters
- Orthogonal Columns, Diagonal and Symmetric Matrices
- Estimation of model R2, Adjusted R2, (?? or r) to assess data efficacy
- Coefficient of Variation (CV)
- Multicollinearity and Variance Inflation Factors
Review of: What Every Engineer Should Know About Regression Analyses
The best overview of the subject material I have found to date
Was this helpful?
391-What Every Engineer Should Know About Regression Analysis
Overall a good review of regression basics. There were a few typos in the test, but everything was promptly corrected by Prof Okogbaa and SunCam.
Was this helpful?
Other Courses by O. Geoffrey Okogbaa, Ph.D., PE
Local Copy of Course Preview