MACHINE LEARNING | DATA | PYTHON

Linear Regression versus Poisson Regression

A guide to understanding the right tool for your data

Image by author (combination of AI and manual adjustments).

Machine Learning and Linear Regression

Perhaps the first machine learning algorithm that we all learn, the linear regression algorithm is surely one of the most fundamental and heavily used techniques — especially when it comes to predictive analytics.

The main benefit of linear regression lies in its simplicity and can be implemented quite easily. And being simple means that the algorithm is also highly interpretable.

In its essence, linear regression attempts to model a given dataset by fitting a linear equation to the observed data points.

As such, the key concept here is linearity. The assumption that there exists a linear relationship between our features (can either be one or more) and some target objective.

This linearity can be represented by an equation of the form:

y = β0 + β1*x1 + β2*x2 + ... + βn*xn + ε

Where:

• y is the dependent variable (the outcome we are trying to predict).
• x1, x2, ..., xn are the independent variables (the predictors).