Linearity and Nonlinearity in Linear Regression | Statistics Tutorial #33 | MarinStatsLectures

Nonlinearity in Linear Regression | Statistics Tutorial #33 | MarinStatsLectures Nonlinearity in Linear Regression: What to do when the relationship between X and Y is not linear and how to transform nonlinear data? Linear Regression in R: https://goo.gl/twYZ4W - More Statistics and R Programming Tutorials: https://goo.gl/4vDQzT ► Linearity between X and Y is an important assumption of linear regression, probably the most important of the assumptions. Regardless of whether your goal is to predict Y, or to estimate the effect of X on Y, linearity is an important and necessary assumption. ► In this video we discuss approaches one can take when the relationship between X and Y is not linear and you have to deal with nonlinearity in linear regression ► Approaches discussed in this lecture include transformations of X and/or Y, polynomial regression (including quadratic terms in the model), converting X to a categorical variable (to a factor), and using a non-linear regression approach. Other approaches aside from these do exist as well. ► We have separate videos discussing how to check the assumption of linearity (and other assumptions), as well as a separate video discussing polynomial regression specifically ✹ Table of Content: 0:00:08 Linear Regression Assumptions 0:02:09 predictive model vs effect size model 0:02:19 Different approaches to address nonlinearities in linear regression model 0:02:31 Transforming the Y variable to address nonlinear data 0:05:02 Pros and Cons of Transforming Y to address nonlinearity 0:05:22 Explaining constant variance or homoscedasticity 0:05:54 transforming Y for predictive model vs effect size model 0:06:58 Transforming the X variable to address nonlinearity in linear regression model 0:07:34 what is Ladder of transformations 0:08:10 Pros and Cons of using ladder of transformations to address nonlinearity 0:09:14 polynomial or a quadratic approach to address non-linearity 0:12:35 Categorizing X (making it a factor) to address nonlinearity 0:14:54 Pros and Cons of categorizing X variable to address non linearity 0:16:25 using a nonlinear regression model to address nonlinearity 0:17:10 Pros and Cons of nonlinear regression model ►►More to Watch : ► Linear Regression in R Series (all videos): https://goo.gl/twYZ4W ► Simple Linear Regression in R https://youtu.be/66z_MRwtFJM ► Checking Linear Regression Assumptions in R https://youtu.be/eTZ4VUZHzxw ► Multiple Linear Regression in R: https://youtu.be/q1RD5ECsSB0 ► Changing a Numeric Variable to Categorical Variable in R https://youtu.be/EWs1Ordh8nI ►Dummy Variables or Indicator Variables for Regression in R https://youtu.be/2s8AwoKZ-UE ► Change Reference/Baseline Category for a Categorical Variable in Regression Model in R https://youtu.be/XJw6xdBYG7c ► Categorical Variables or Factors in Linear Regression in R https://youtu.be/KHTBwTBkCzg and https://youtu.be/ZtBmMhGkxxA ► Multiple Linear Regression with Interaction in R https://youtu.be/8YuuIsoYqsg ► Interpreting Interaction in Linear Regression https://youtu.be/vZUtDJbzFRQ ►Variable Selection in Linear Regression Using Partial F-Test in R https://youtu.be/G_obrpV70QQ ► Polynomial Regression in R https://youtu.be/ZYN0YD7UfK4 Follow MarinStatsLectures Subscribe: https://goo.gl/4vDQzT Visit Website: https://statslectures.com Facebook:https://goo.gl/qYQavS Twitter:https://goo.gl/393AQG Instagram: https://goo.gl/fdPiDn Our Team: Content Creator: Mike Marin (B.Sc., MSc.) Senior Instructor at UBC. Producer: Ladan Hamadani (B.Sc., BA., MPH) These videos are created by #marinstatslectures to support a course at The University of British Columbia (UBC) (#SPPH400: #IntroductoryStatistics for Health Science Research), although we make all videos available to the public for free. Thanks for watching! Have fun and remember that statistics is almost as beautiful as a unicorn!