Regression Modeling in R
Go to main | Course Page
Section 1: Overview
- Course Overview
- How to Get Queries Resolved
- Download Resources
Section 2: Introduction to Linear Regression
- What is Linear Regression
- Understanding with Practical Examples
- Types of Linear Regression
- Examples of Industrial Applications
- Statistical Modeling vs Machine Learning
Section 3: Graphical Understanding
- Graphical Understanding
- Formulate Line of Best Fit
- Course Review
Section 4: Building and Interpreting Linear Regression
- Linear Regression from Scratch using Formula
- R-Squared Explained
- Pre-Model Build Analysis
- Building and Interpreting Linear Regression Models
Section 5: Measures of Goodness of Fit
- Problem with R-Squared
- Adjusted R-Squared
- F-Statistic, AIC, BIC
- R-Demo
- Assumptions
- Course Review
Section 6: Project - Linear Regression
- Problem Statement
- Handling Missing Values
- Outlier Analysis
- Graphical and Statistical Analysis
- Building Linear Regression
- Good Model
- Evaluation Measures
- Tips to Improve Model Accuracy
- Need for Cross Validation
- Cross Validation Approaches
- Variable Transformations and Interactions
- Variable Transformations and Interactions
- Cooks Distance for Influential Points
- Cooks Distance Demo
- BoxCox and YeoJohnson Transformations
- Residual Analysis
- Overcoming Heteroscedasticity
- Stepwise Regression for Model Search
- Best Subsets Model Search
Section 7: Gradient Descent
- What Exactly is Gradient Descent
- How Gradient Descent Learns
- Comparing Types of Gradient Descent
- Stopping Criteria and Scaling
- Types of Gradient Descent
- Course Review
Section 8: Introduction to Logistic Regression
- Introduction to Logistic Regression
- One vs Rest Strategy
- Use Case Examples
Section 9: Understanding the Math
- Math behind Logistic Regression Part 1
- Math behind Logistic Regression Part 2
- Why Negative Logloss
- Course Review
Section 10: Project - Logistic Regression
- Problem Statement Marketing
- Demo - EDA for Logit
- Building Logistic Regression Model
- McFadden's R-Squared
- Confusion Matrix and Evaluation Metrics
- Precision Recall Curve
- Optimal Cutoff Score
- The ROC Curve
- KS Statistic and Gain Curve
- Concordant and Discordant pairs
- Approaches to Handle Class Imbalance
- Cost Sensitive Learning
- Oversampling
- Hybrid Sampling