The linear regression is a method used to model the relationship between a dependent variable and one or more independent variables by fitting a linear equation to observed data. Building a regression model with stats packages has become straightforward. However, interpreting the software output and building a good model are no simple tasks. Learn the essentials of model-building, goodness-of-fit tools, model validation & common pitfalls.
In this course, the construction of regression models is covered with the help of real case studies.
Course Outline
Session 1: Simple Linear Regression – SLR
- General principle
- Model specification
- Estimating model coefficients using the data
- Interpretation & significance testing of model coefficients
- Goodness-of-fit & validation tools
- Issues in regression: Outliers, Influential observations, etc.
Session 2: Multiple Linear Regression – MLR
- Parcimonious modelling: Variable selection
- Multicollinearity
- Prediction (interpolation)
- Dangers of extrapolation
- Model validation
Course Duration
The recommended course duration is 2 online sessions.
Target Audience
This module is intended for anyone who collects data and makes decisions based on this data. Knowledge of the regression methods covered in this training will be particularly useful for people who have to relate or predict a variable to a single or a set of explanatory variables.
Participants must know the essential tools of descriptive statistics and inferential statistics – mean, standard deviation, standard error, median, graphical tools such as histograms, box plots, hypothesis tests, confidence intervals, etc. That is to say, have followed the training Fundamental Statistical Tools or have an equivalent level.