Machine Learning: 2 Books in 1: Machine Learning for Beginners, Machine Learning Mathematics. An Introduction Guide to Understand Data Science Through the Business Application

“Multiple Linear Regression" tends to be the most common form of
“regression" technique used in data science and the majority of statistical
tasks. Just like the “linear regression” technique, there will be an output
variable "Y" in "multiple linear regression". However, the distinction now is
that we're going to have numerous “X” or independent variables
generating predictions for "Y".
For instance, a model developed for predicting the cost of housing in
Washington DC will be driven by "multiple linear regression" technique.
The cost of housing in Washington DC will be the "Y" or dependent
variable for the model. "X" or the independent variables for this model will
include data points such as vicinity to public transport, schooling district,
square footage, and some rooms, which will eventually determine the
market price of the housing.

---

The mathematical equation for this model can be written as below:
“housing_price = β0 + β1 sq_foot + β2 dist_transport + β3 num_rooms”
“Polynomial regression” - Our models developed a straight line in the last
two types of "regression" techniques. This straight line is a result of the
connection between "X" and "Y" which is "linear" and does not alter the
influence "X" has on "Y" as the changing values of "X". Our model will
lead in a row with a curve in “polynomial regression”.
If we attempted to fit a graph with non-linear features using "linear
regression", it would not yield the best fit line for the non-linear features.
For instance, the graph on the left is shown in the picture below has the
scatter plot depicting an upward trend, but with a curve. A straight line does
not operate in this situation. Instead, we will generate a line with a curve to
match the curve in our data with a polynomial regression, like the chart on
the right shown in the picture below. The equation of a polynomial will
appear like the linear equation, the distinction being that one or more of
the "X" variables will be linked to some polynomial expression. For
instance,
“Y = mX2+b”

---

Another significant “regression” technique for data researchers is "Support