Duration: 6 weeks
Target Audience: R users with statistics knowledge (distributions, confidence intervals, central tendency, dispersion etc) and basic data skills, aiming to master regression techniques.
Week 1: Foundations of Regression Analysis
- Review of statistical concepts: correlation, linearity, assumptions.
- Overview of regression types: SLR, MLR, Logistic.
- Assignment: Explore a dataset and check for linear relationships.
Week 2: Simple Linear Regression (SLR) in R
- Building SLR models using lm().
- Assessing model fit: R-squared, residuals.
- Assignment: Fit an SLR model and interpret coefficients.
Week 3: Multiple Linear Regression (MLR) in R
- Adding multiple predictors, handling multicollinearity.
- Model diagnostics: VIF, residual plots.
- Assignment: Build an MLR model and evaluate its performance.
Week 4: Logistic Regression in R
- Fitting logistic models with glm() for binary outcomes.
- Model evaluation: confusion matrix, ROC curve.
- Assignment: Fit a logistic regression model for a binary outcome.
Week 5: Model Selection and Practical Applications
- Techniques: stepwise regression, cross-validation.
- Applying regression to real-world data with visualizations (ggplot2).
- Assignment: Perform model selection and create a report with results.
Week 6: Capstone Project
- Develop a regression analysis project (e.g., predicting sales or patient outcomes).
- Use SLR, MLR, or Logistic models and present findings.
- Final feedback session.
