Credit Default Prediction and Stock Market Risk Analysis

Course: Finance and Risk Analytics

Overview

This project is divided into two main parts:

• Part A: Credit Default Prediction aims to predict the likelihood of companies defaulting on their debt obligations using financial data and machine learning models.
• Part B: Stock Market Risk Analysis focuses on assessing the volatility and performance of Indian stocks, providing valuable insights into market risk for investors.

These analyses serve as essential tools for financial institutions, investors, and other stakeholders to make data-driven decisions related to creditworthiness and investment strategies.

---

Project Objectives

• Part A: Develop a predictive model to assess company credit default risk using historical financial data.
• Part B: Evaluate the volatility and returns of a portfolio of stocks to understand market risks and optimize investment strategies.

---

Project Structure

• Data: Contains the datasets used for the analysis.
• Notebooks: Contains Jupyter notebooks with detailed analysis and code for both parts.
• Reports: Includes the comprehensive business report summarizing insights and recommendations from the analyses.

---

Skills & Tools Utilized

• Data Preprocessing & Cleaning
• Exploratory Data Analysis (EDA)
• Predictive Modeling (Logistic Regression, Random Forest)
• Statistical Analysis (Mean, Standard Deviation)
• Financial Metrics & Market Risk Assessment
• Libraries: pandas, numpy, sklearn, matplotlib, seaborn

---

Part A: Credit Default Prediction

Problem Statement

Predict a company's ability to meet debt obligations based on financial metrics, using machine learning models to provide actionable insights for risk management.

Approach

1. Data Understanding & Preparation:
   • Loaded and explored financial data for various companies.
   • Preprocessed data by handling missing values and treating outliers.
2. Modeling:
   • Built and evaluated predictive models (Logistic Regression, Random Forest).
   • Enhanced performance through feature engineering, multicollinearity treatment, and hyperparameter tuning.
3. Insights:
   • Key features influencing credit default risk include retained earnings, net profit ratios, and total debt.
   • Developed risk mitigation strategies based on predictive outputs.

---

Part B: Stock Market Risk Analysis

Problem Statement

Assess the risk and performance of a portfolio of Indian stocks using historical price data, focusing on volatility and returns to inform investment strategies.

Approach

1. Data Exploration & Preparation:
   • Analyzed weekly stock price data for five Indian companies.
   • Cleaned and prepared data for statistical analysis.
2. Risk and Return Analysis:
   • Computed historical returns and analyzed mean and standard deviation for each stock.
   • Visualized stock trends and assessed risk-reward trade-offs.
3. Insights:
   • Stocks with higher volatility posed greater risks; specific stocks demonstrated consistent performance.
   • Balanced investment strategies are recommended based on risk tolerance.

---

How to Run

• Clone this repository.
• Open and run FRA_Main_Project_Part_A.ipynb and FRA_Main_Project_Part_B.ipynb using Jupyter Notebook or Google Colab.

---

Key Insights & Recommendations

Part A

• Focus on key financial metrics influencing creditworthiness to enhance risk assessment.
• Leverage predictive models to anticipate and mitigate potential defaults.

Part B

• Consider diversification strategies to manage portfolio risk.
• Use volatility measures to balance risk and return effectively.

---

Learnings

• Mastered data cleaning, feature engineering, predictive modeling, and financial risk analysis techniques.
• Applied statistical measures and visualization techniques to inform financial decision-making.