grayoj / comex

Linear regression to perform predictive analysis on individual's income.

Linear Regression Model

---

In this project (jupyter notebook) we are using a SickitLearn model, to perform linear regression to make income predictions with the dataset made available in the repository.

Linear regression, is a supervised machine learning model which implements a linear relationship between an independent and dependent variable. It is a statistical method of making predictive analysis. We can illustrate linear regression by the below formula:

$y=a$_$0$$+a$_$1$$x+ ε$

Where

• $Y$ represents = The Dependent Variable (Target Variable)
• $X$ represents = The Independent Variable (Predictor Variable)
• $a0$ represents = The intercept of the line (Gives an additional degree of freedom)
• $a1$ represents = The Linear regression coefficient (Scale factor for each input value)
• $ε$ represents = The Random error

Preparing Data For Linear Regression

---

1. Linear Assumption.
2. Eliminate Noise (Carry out data cleaning)
3. Eliminate Collinearity.
4. Rescale the provided inputs.

Requirements

These are the tools, environment variables and libraries used in the project.

1. Jupyter Notebook.
2. Python 3+
3. SickitLearn.
4. Pandas.
5. Numpy.
6. Data Set (Already Provided)

Installation

To use this project, you need to clone the repository.

> git clone https://github.com/grayoj/income-prediction.git

After you finish cloning, install these modules using Pip

Install Sklearn

> pip install sklearn

Install Pandas to read datasets (CSV)

> pip install pandas

Install NumPy

> pip install numpy

Ensure you have Jupyter Notebook installed, which you could easily set up if you use VsCode

Try out the notebook!

Any suggestions, feedback and help, as well as improvements are all welcome and will be appreciated. Contact! Mail me, here.

• MIT License.