xre22zax / Linear-regression-website-sample

Linear regression and prediction for website

Linear Regression Analysis of Website Usage Time

This repository contains Python code that explores factors influencing time spent on a website using linear regression.

Key Findings

• Age and time spent on the website are positively correlated.
• Chrome users tend to spend longer on the website than Safari users.

Files

• website.csv: Contains data on user age, browser type, and time spent on the website (in seconds).
• linear_regression.py: Python script for analysis, including:
  • Loading and exploring data
  • Building and fitting linear regression models
  • Visualizing results
  • Making predictions

Dependencies

• pandas
• numpy
• matplotlib.pyplot
• statsmodels.api

Usage

1. Install dependencies: pip install pandas numpy matplotlib statsmodels
2. Run the script: python linear_regression.py

Key Code Snippets

# Load data
website = pd.read_csv('website.csv')

# Model 1: Time vs. Age
model = sm.OLS.from_formula('time_seconds ~ age', website)
results = model.fit()

# Model 2: Time vs. Browser
model = sm.OLS.from_formula('time_seconds ~ browser', website)
results = model.fit()

# Prediction for 40-year-old
pred40 = results.params[0] + results.params[1]*40

Contributing

Feel free to submit issues or pull requests for improvements or additions.

---

Author

Reza Sadeghi