Practical Introduction to Linear Regression using Excel

This course takes the participants through the principles of Ordinary Least Squares (OLS) and linear regressions. 

Participants will explore relationships between variables and build models using Excel functions and features such as SLOPE, INTERCEPT, LINEST, Solver and the correlation and regression tools in the Analysis Toolpak.

Advanced topics include multivariate regression and statistical significance testing using t-tests and F-tests.

Guided exercises support the students to ensure practical understanding and immediate applicability of the concepts.

Introduction to regression

• Exploring relationships: scatter plots, variance, covariance, and Pearson’s correlation coefficient.
• Visualizing relationship : scatter plots.
• The functional form and building a linear regression model: terminology, slope, intercept, residuals, and how to determine the best fit line

Linear regression with one predictor

• The basic Excel tools: Trendline, Data Tables, Solver, and the functions SLOPE, INTERCEPT, LINEST
• Determining goodness of fit: explained and unexplained variance, coefficient of determination (R2).
• Analysing the residuals: verifying normality and absence of correlation in the residuals, and use of QQ plots. 

Statistical inference and statistical significance

• The t-statistics, F-statistic, and evaluating model significance
• Interpreting the Analysis Toolpak output
• Testing the residuals: Jarque-Bera, Breusch-Pagan and Durbin-Watson tests.

Transforming a nonlinear relationship into a linear model.

• Properties of logarithms
• Logarithmic transformations: LOG-LINEAR models
• The power and exponential laws

Regression with more than one predictor

• Multivariate regression
• Detecting multicollinearity (correlation matrix and variance Inflation Factor)
• Interpreting model results: Partial effects
• Assessing goodness of fit: Adjusted R²
• Model significance and model comparisons (F-test)

• State the principles of linear regression and the requirements for Ordinary Least Squares
• Identify the various methods available in Excel for linear regression and when to use each approach
• Apply Excel’s functions to produce linear models using one or multiple predictors
• Calculate goodness of fit parameters and discuss model usability
• Assess if a model is useful for inference using statistical significance testing
• Apply logarithmic transformations when relationships appear non-linear
• Communicate regression results clearly and confidently

Classroom delivery coupled with hands-on applications via case examples using regression template files and interactive sessions.