Linear Regression and Regularisation

Go to main | Course Page

Section 1: Introduction to Linear Regression

  1. What is Linear Regression
  2. How to Get Queries Resolved
  3. Download Resources
  4. Regression Equation and Terminologies
  5. Time Series Regression
  6. Formula for Coefficients
  7. Coefficients Computation from Scratch

Section 2: Regression Algorithm from scratch: Gradient Descent

  1. What is Gradient Descent
  2. Math behind Gradient Descent
  3. Stochastic and Minibatch
  4. Gradient Descent from Scratch - Coding
  5. Course Review

Section 3: Model Building and Concepts Part- 1

  1. Problem Description and Data
  2. Describe the Data
  3. EDA Part 1 Understanding Sales
  4. EDA Part 2 Graphical Analysis
  5. Missing Values and Outlier
  6. Outliers and mahalanobis distance
  7. Building and Interpreting Linear Regression
  8. R-Squared Intuition
  9. Adjusted R Squared and F Statistic

Section 4: Model Building and Concepts Part- 2

  1. Assumptions of OLS - Part 1
  2. Assumptions of OLS - Part 2
  3. Assumptions of OLS - Part 3
  4. VIF
  5. Durbin Watson statistic and Condition Number
  6. Multicollinearity
  7. How to check and rectify Heteroscedasticity
  8. BoxCox Transform
  9. The Other Way
  10. Model Improvement Tactic Demo- Cooks D

Section 5: Model Selection Approaches

  1. Evaluating Regression Models
  2. Cross Validation Approaches
  3. Need for Holdout Sample
  4. Tactic 1 - LR Model Building
  5. Tactic 2 - Backward Building Workout
  6. Stepwise and Best Subsets
  7. RFE and Caveats
  8. RFE Demo

Section 6: Regularization Modelling Approaches

  1. Bias Variance Tradeoff
  2. Ridge Regression
  3. Grid Search
  4. LASSO Regression
  5. ElasticNet
  6. Ridge Regularization Code Demo
  7. Why weights of L1 regularization reach zero but not L2

Section 7: Outlier Resistant and Advanced Regression

  1. RANSAC for Outliers Resistant Models
  2. Theil-Sen Regression
  3. Time Series Regression
  4. Retraining Models
  5. How to Identify if Data Drift occurred
  6. Quantile Regression
  7. Robust Regression with Huber Loss
  8. Advanced Regression Code

Section 8: Maximum Estimation for Linear Regression

  1. Linear Regression with Maximum Likelihood Estimation
  2. Assignment
Report abuse