Regression analysis is a useful mechanism for estimating the relationship between a dependent variable and one or more independent variables. It is widely used in forecasting and has become an important machine learning tool. It becomes crucial for someone starting in machine learning to understand how regression analysis works.

Let us look at a few resources available online to get started with regression analysis. 

MachineHack Bootcamp

MachineHack, a popular platform for data scientists and AI practitioners provides courses on regression in the form of bootcamps. Bootcamps are pocket courses for all who aspire to become data scientists, data engineers and machine learning developers. It has divided the course into small portions that can be covered at your own pace.Some popular courses in this bootcamp on regression include simple linear regression, multiple linear regression, polynomial Regression for non-linear modelling, decision tree for regression. 

For more details, click here.

Course on Regression Analysis by NPTEL 

The free course is conducted by National Programme on Technology Enhanced Learning (NPTEL) and coordinated by IIT Kharagpur and taught by Dr. Soumen Maity, (then) faculty at IIT Kharagpur. It has video lectures and transcripts that can also be downloaded for better understanding. It starts with explaining the concept of regression across several videos, moves to multiple linear regression and how to choose the best regression model as per your needs. It then moves into the concept of multicollinearity, model adequacy checking and testing for influential observations. The course also introduces the concepts of polynomial regression models, generalised linear models, non-linear estimation, regression models with autocorrelated errors, measurement errors and calibration problems, among others.

The course can be accessed here.

Deep Learning Foundation: Linear Regression and Statistics by Udemy

This course will introduce students to the mathematics behind R-Squared, linear regression and variance inflation factor (VIF), the statistical background of linear regression and assumptions, and assumptions of linear regression hypothesis testing. They will also be taught to program their own version of a linear regression model in Python and implement it appropriately to data science problems. The course will also cover writing codes for T-Test, Z-Test and Chi-Squared Test in Python. Currently, it is priced at Rs 1280.

For more details, click here.

Linear Regression and Modeling: by Duke University and Coursera

As per Coursera, this course will teach students simple and multiple linear regression models. They will be taught the fundamental theory behind linear regression, and with the help of examples, they can learn to fit, examine, and use regression models to analyse relationships between multiple variables. They will also be introduced to R and RStudio.

For more details, click here.

Regression Analysis: an Introduction by UCL

This course will provide students with an overview of regression types and the application of multiple linear regression. It will look at the theory that forms the foundation of regression analysis (focus on linear regression), formulation and interpretation of linear regression models.

There will be an optional extra two days for hands-on use of a statistical package (SPSS). It is completely online and suitable for anyone who needs an understanding of introductory-level regression analysis.

For more information, click here.

Data Science: Linear Regression by Harvard University and edX 

This course by Harvard University and edX will teach students about the origin of linear regression by Galton, how to detect confounding and how to look at relationships between variables by using linear regression in R. Students can audit the course for free but to add a certificate, they need to pay $99. The course is self-paced and takes around eight weeks to complete by putting in one to two hours per week.
For more details, click here.