Bayesian Analysis of Iris Dataset

Overview

This project uses Bayesian statistical modeling to analyze the sepal_width feature of the Iris dataset. The project compares different models (Normal and Student's-T) and evaluates the robustness of each model in handling the data distribution. It further extends to group comparisons across three species of the Iris dataset (Setosa, Versicolor, and Virginica) using Bayesian modeling and statistical metrics.

Objectives

• Estimate the prior and likelihood of the sepal_width feature using both Normal and Student's-T distributions.
• Compare the posterior results of these models to determine which is more robust.
• Group the data by species and analyze sepal_width using Bayesian models.
• Apply group comparison techniques, such as Cohen’s d and Probability of Superiority, to quantify differences between species.

Requirements

The following libraries are required to run this project:

• pymc
• arviz
• pandas
• numpy
• matplotlib
• seaborn
• scipy

Install them using:

pip install pymc arviz pandas numpy matplotlib seaborn scipy