Course Content

Course Content


More than just fitting a linear line through a cloud of data points is involved in linear regression analysis. It has three stages: (1) examining the data for correlation and directionality, (2) estimating the model, i.e. fitting the line, and (3) evaluating the model's validity and utility.

A linear regression model is predicated on four assumptions: Linearity refers to the relationship between X and the mean of Y. Homoscedasticity: For every value of X, the variance of the residual is the same. Independent observations: Observations are not reliant on one another.

y = b0 + b1*x + e is the linear regression mathematical formula, where b0 and b1 are known as the regression beta coefficients or parameters: The anticipated value when x = 0 is b0, which is the intercept of the regression line. The slope of the regression line is b1.

Linear Regression is a supervised machine learning technique with a continuous and constant slope projected output. Rather than aiming to classify data into categories, it's used to predict values within a continuous range (e.g. sales, price) (e.g. cat, dog).

These are some considerations to make when selecting a linear model:

  • Comparing linear models for the same dataset is the only way to go.
  • Look for a model with a high R2 adjusted.
  • Ascertain that the residuals in this model are evenly distributed around zero.
  • Ascertain that the model's faults are contained within a narrow bandwidth.

Recommended Courses

Share With Friend

Have a friend to whom you would want to share this course?

Download LearnVern App

App Preview Image
App QR Code Image
Code Scan or Download the app
Google Play Store
Apple App Store
598K+ Downloads
App Download Section Circle 1
4.57 Avg. Ratings
App Download Section Circle 2
15K+ Reviews
App Download Section Circle 3
  • Learn anywhere on the go
  • Get regular updates about your enrolled or new courses
  • Share content with your friends
  • Evaluate your progress through practice tests
  • No internet connection needed
  • Enroll for the webinar and join at the time of the webinar from anywhere