Welcome to the Insurance Risk Analysis Project! This repository contains analyses aimed at optimizing marketing strategies for AlphaCare Insurance Solutions by examining historical car insurance claim data.
The goal of this project is to perform comprehensive data analysis, including exploratory data analysis (EDA), hypothesis testing, machine learning modeling, and more. Each task is organized into separate branches for clarity and better management of changes.
-
- Exploratory Data Analysis (EDA)
- Data summarization and descriptive statistics
- Data quality assessment and missing values check
- Univariate and bivariate analyses
- Visualizations capturing key insights
- Exploratory Data Analysis (EDA)
-
- Data Version Control (DVC)
- Implementation of DVC for dataset tracking
- Setting up local storage and managing data versions
- Data Version Control (DVC)
-
- A/B Hypothesis Testing
- Testing hypotheses regarding risk differences across various demographics
- Statistical analysis and reporting of results
- A/B Hypothesis Testing
-
- Statistical Modeling
- Preparing data for modeling and handling missing values
- Building and evaluating multiple models (Linear Regression, Random Forests, XGBoost)
- Analyzing feature importance and model performance
- Statistical Modeling
To explore the project:
- Clone this repository:
git clone https://github.com/AschalewMathewosDamtew/insurance-risk-analysis.git
- Navigate into the cloned repository:
cd insurance-risk-analysis
- Checkout the desired branch (replace {branch-name} with the actual branch name, like task-1-eda-visualization, etc.):
git checkout {branch-name}