Join me on a journey through my Python Portfolio, where I showcase my projects and expertise in data analysis and visualization. With a focus on leveraging the power of Python, I present a collection of diverse projects that demonstrate my skills in extracting insights and generating value from data.
| Project Link | Project Focus | Project Description |
|---|---|---|
| Artificial Intelligence / Analytics Capstone - Isolation Forest for Outlier Detection in Insurance Data | Outlier Detection, Dimensionality Reduction, Data Visualization | In this project, I utilized the Isolation Forest algorithm to detect outliers in insurance data and uncover potential cases of adverse selection. I applied PCA for dimensionality reduction and generated a heatmap to visualize the outliers. Moreover, I integrated a demographic dataset to provide context and created visualizations to explore the demographic characteristics of the identified anti-selector users. |
| Artificial Intelligence / Analytics Capstone - Apriori Algorithm for Adverse Selection Detection in Insurance | Association Rule Mining, Adverse Selection Detection | For this project, I detect adverse selection in insurance by utilizing the Apriori algorithm. I identified frequent item sets and generated association rules to uncover potential cases of adverse selection. By combining these findings with demographic data, I provided insights into the characteristics and behaviours of anti-selector users, helping to address and mitigate adverse selection in the insurance industry. |
| Advanced Data Analytics - Wine Quality Classification using Decision Trees | Classification, Decision Trees | In this project, I applied a decision tree algorithm to classify wine quality based on various attributes. I evaluated the performance of the decision tree using accuracy metrics and visualized the decision-making process to understand the important features of wine quality classification. |
| Exploratory Data Analysis of Grocery Transaction Dataset | Data Preparation, Descriptive Statistics, Data Visualization | Task 1 of the Cognizant Virtual Internship. In this project, I analyzed a grocery transaction dataset to uncover its structure and statistical properties. I aimed to gain valuable insights for optimizing stock management practices through thorough data examination, descriptive statistical analysis, and insightful visualizations. The analysis also identified areas for further exploration and potential data requirements to enhance the understanding of the dataset. |
| Predictive Model for Hourly Product Stock Level Forecasting | Data Modeling, Forecasting, Feature Importance | Task 3 of the Cognizant Virtual Internship. In this project, I built a predictive model to forecast product stock levels on an hourly basis. It involved data preparation, creating relevant features, training a RandomForestRegressor model, and evaluating its performance. Additionally, I visualized the importance of features using a bar plot. |
| Python Module Development for Machine Learning Production | Model Deployment, Python Module Development, Data Processing | Task 4 of the Cognizant Virtual Internship. In this project, I transformed the code from the previous task into a functional Python module capable of reading a CSV file and performing data modelling. The module enabled efficient data processing and accurate prediction of stock levels. It also included comprehensive comments and documentation to ensure usability by the machine learning engineering team. |
| Statistical Analysis of Financial Data | Statistical Analysis, Financial Insights | The first assignment of the applied portfolio management subject. Here I conducted a statistical analysis of financial data. I performed correlation and quantile analysis and calculated the information coefficient. The project explored factors influencing investment strategies, such as quality, profitability, growth, and safety. |
| Portfolio Optimization and Backtesting | Portfolio Management, Backtesting, Performance Metrics | The second assignment of the applied portfolio management subject. Here I implemented portfolio management strategies and backtested the Quality factor. I optimized strategies through parameter exploration and calculated performance metrics. The project also included stability analysis, walk-forward modelling, and backtesting of a combined factor with profit, growth, and safety. |
| Financial Data Exploration and Analysis | Data Analysis, Financial Data Exploration | Week 3 Coding Exercise for the applied portfolio management subject. Here I loaded financial data and calculated averages and correlations. I compared market returns and used fundamental programming techniques to extract valuable insights on key metrics and trends within financial datasets. |
| Momentum Indicators in Financial Data Analysis | Data Analysis, Momentum Strategies | Week 5 Coding Exercise for the applied portfolio management subject. Here I explored momentum indicators and their analysis. I applied momentum strategies in the financial domain by calculating information coefficients and examining performance metrics. |
| Quantile Analysis and Strategy Evaluation in Portfolio Management | Data Analysis, Portfolio Management, Quantile Analysis | Week 6 Coding Exercise for the applied portfolio management subject. Here I performed quantile analysis and strategy evaluation. I divided stocks into groups based on factors like momentum and earnings-to-price ratios. I analyzed their performance metrics, highlighting the importance of portfolio diversification and the impact of different factors on investment strategies. |
| Statistical Tests and Hypothesis Testing in Financial Data Analysis | Data Analysis, Statistical Testing, Hypothesis Testing | Week 7 Coding Exercise for the applied portfolio management subject. Here I analyzed financial data and conducted statistical tests. I performed tasks such as merging datasets, performing t-tests, and backtesting investment strategies. The project demonstrated the application of statistical analysis and hypothesis testing in the financial domain. |
| Predictive Modeling with Decision Tree Regression for Stock Returns | Predictive Modeling, Machine Learning, Stock Return Forecasting | Week 11 Coding Exercise for the applied portfolio management subject. Here I performed predictive modelling using decision tree regression. I applied machine learning techniques to forecast future stock returns. The project included training a model on historical financial data, making predictions, and evaluating performance metrics. I emphasized feature selection and model evaluation for accurate predictions. |