### Course Description

In this course you will learn to apply Least Squares regression and it's assumptions to real world data. Then we'll improve on that algorithm with Penalized Regression and non-parametric Kernel methods
- Understanding basic statistical modeling and assumptions
- Build & Evaluate supervised linear models using:
Least squares
Penalized least squares
Non-parametric methods
- Model selection and fit on real world applications:
Insurance
Healthcare
etc.
- Code samples
https://bitbucket.org/arthuranalytics/experfy_courses/src/master/
Topics:
Introduction – Supervised Learning and ML
ML Statistics – Understanding Assumptions
Least Squares Regression – The ML workhorse
Linear Model Evaluation– Assess performance
Penalized Regression (L1/L2) – Optimization
Kernel Methods (SVM) – Other Distributions
Real World Applications

#### What am I going to get from this course?

Students will learn and understand the applications of the following algorithms:

- Least squares
- Penalized least squares
- Non-parametric methods
- Model selection and fit on real world applications:
- Insurance
- Healthcare
- etc.

### Prerequisites and Target Audience

#### What will students need to know or do before starting this course?

Programming Knowledge:

- Python (preferred)
- Scikit-Learn exposure
- Basic Statistics
- Exposure to Linear Algebra
- Completed: Course 1
- Machine Learning for Predictive Data Analytics

#### Who should take this course? Who should not?

This course is best for practitioners that have exposure to basic statistics and are interested in machine learning.

Managers who would like exposure to linear algorithms would also benefit from this material.

### Curriculum

Introduction to Supervised Linear Regression Course
- Prerequisites
- Skills gained

#### Module 2: Introduction to Machine Learning and Supervised Regression

Lecture 2
Introduction to Machine Learning and Supervised Regression

- Discuss the overall AI ecosystem and how Machine Learning (ML) is part of that ecosystem.
- Understand the 3 different types of algorithms that make up ML
- Provide some intuition for why functions and optimizations are important in ML.
- Differences between Statistical and ML approaches to supervised linear regression.

Quiz 1
Module 2 - ML and Supervised Regression

#### Module 3: Machine Learning - Understanding assuptions

Lecture 3
Machine Learning - Understanding Assuptions

- Survey the statistical concepts important to understanding Linear Algorithms.
- Design of experiments.
- Conducting experiments.
- Understand the difference between linear and non-linear functions.

Quiz 2
Module 3 - Linear Regression Assumptions

#### Module 4: Least Squares Regression - Ordinary Regression

Lecture 4
Least Squares Regression - Ordinary Regression

Develop the simple linear regression algorithm.
Understand the basic linear regression assumptions.
Learn to identify when assumption violations occur.
Understand how to evaluate model output.

Quiz 3
Module 4 - Simple Regression

#### Module 5: Least Squares Regression - Multiple Regression

Lecture 5
Least Squares Regression - Multiple Regression

Extend the Least Squares algorithm to multiple dimensions
Explore data to understand variable importance
Prepare data for multiple regression
Optimizing between Bias and Variance

Quiz 4
Module 5 - Multiple Regression

#### Module 6: Penalized Regression - L1/L2 Optimization

Lecture 6
Penalized Regression - L1/L2 Optimization

Understand motivation behind penalized regression
Optimize L1 Regression (Lasso) parameters
Optimize L2 Regression (Ridge) parameters
Combine the L1/L2 penalties (Elastic Net)
Understand the difference and trade offs between Subset Selection and Shrinkage
Optimize hyper-parameters with Cross-Validation

Quiz 5
Module 6 - Penalized Regression

#### Module 7: Kernel Methods - Support Vector Machines

Lecture 7
Kernel Methods - Support Vector Machines

Understand theory and motivation behind kernel methods.
Derive a basic kernel and use the kernel trick.
Build a support vector classifier.
Extend to regression with support vector machine.
Optimize parameters with Cross validation and Grid Search

Quiz 6
Module 7 - Support Vector Machines

#### Module 8: Kernel Methods - Gaussian Process Regression

Lecture 8
Kernel Methods - Gaussian Process Regression

Understand multivariate distributions and non-parametric regression.
Use Bayesian probability with joint probabilities.
Develop theory behind Gaussian Process Regression.
Optimize kernels and hyper-parameters.

Quiz 7
Module 8 - Gaussian Process Regression

#### Module 9: Summary and Real World Applications

Lecture 9
Summary and Real World Applications

Review Supervised Linear Regression topics.
Perform Linear regression on real world data.

Download real world dataset and perform regressions and validations to minimize mean square error on the predictions.