You are on page 1of 5

Polynomial Linear Regression

Simple Linear Regression:


Multiple Linear Regression:
Polynomial Linear Regression
Simple Linear Regression:
Multiple Linear Regression:

• When the distribution of the data is of the above form, then we cannot use
simple or multiple linear regression as data distribution is not of linear form.
• We will not be able to fit it properly through them. In such kind of scenario,
we use polynomial linear regression
Polynomial Linear Regression

• If X & Y are two features where X is independent and Y is dependent, and we


have checked that we cannot fit the data through linear regression, then we
try it for various degrees.
• We first check for degree 2, and if we are able to fit properly, then good;
otherwise, we choose degree 3.
Polynomial Linear Regression
• Selection of degree value is hyperparameter.
• If we choose a smaller value of the degree, then it may turn into underfitting.
• Similarly, if we choose a larger value of degree, then it may turn into
overfitting.
Polynomial Linear Regression

• Polynomial regression is called polynomial linear regression as the


relationship between the dependent variable and the unknown parameters is
still linear, i.e., the degree of unknown parameters is still one.
• Polynomial regression is a refinement for simple linear regression to fit the
data points that are in curved shape rather than a straight-line shape. So, it
can be concluded that it is used for nonlinear data.

You might also like