Machine learning basics (part 5)

In this part, let’s discuss multiple regression a bit (check linear model basics in part 2) and then jump right to general linear model.

Multiple Regression

Calculating p-value is the exact same as before, only this time we make modification with “pmean”.

We can also compare simple linear regression with multiple regression and see if it’s worth it to fit into multiple regression with p-value as in following equation:

If the difference in R² values between simple and multiple regressions is “big” and p-value is “small”, then adding Tail Length to the model is worth the trouble.

General Linear Model (GLM)

A general linear model is statistical linear model that generalizes regression to multiple explanatory variables. These explanatory variables can be continuous or categorical.

The term “general” linear model (GLM) usually refers to conventional linear regression models for a continuous response variable given continuous and/or categorical predictors. It includes multiple linear regression, as well as ANOVA and ANCOVA (with fixed effects only).

For a general linear model, we have:

• Explanatory: one or more numerical or categorical variables
• Response: single numerical variable

General linear models: types

We can do statistical modelling with general linear model in R or other coding tool.