Machine Learning & Deep Learning Algorithms (2024-09 – 2024-12)
- Built and evaluated supervised machine learning algorithms (including Regression, Classification, and Neural Networks) to automate data pattern recognition and enhance analytical efficiency.
- Optimized model performance using cross-validation and error diagnostics in R and Python to ensure baseline algorithm reliability
Bayesian Inference & Data Simulation (2024-02 – 2024-06)
- Implemented Bayesian Inference frameworks to continuously update probability metrics as modern data streams are on boarded, maximizing baseline forecasting reliability.
- Utilized Markov Chain Monte Carlo (MCMC) methods within R/Python environments to simulate complex parameters and validate research hypotheses.
Time Series Modeling & Predictive Forecasting Project (2023-01 – 2023-06)
- Modeled complex time-series data using advanced ARIMA frameworks, conducting rigorous stationarity testing (ADF) and residual diagnostics to ensure extreme predictive accuracy.
- Generated robust future trend projections and systematically evaluated prediction errors using accuracy metrics (MAPE/RMSE) to deliver data-driven insights.
Multivariate Data Analysis & Dimension Reduction (2023-01 – 2023-06)
- Applied advanced dimensionality reduction techniques including Principal Component Analysis (PCA) and Factor Analysis to process high-dimensional multivariate datasets, eliminating multi-collinearity.
- Interpreted complex correlation structures across multiple data variables to simplify data structures for strategic reporting.
Probability Theory & Stochastic Modeling (2022-08 – 2022-12)
- Formulated conditional and joint probability distributions to quantify uncertainty and model complex stochastic systems under real-world operational variables.
- Calculated expected values and risk probabilities to assist in computational testing and minimize experimental errors
Advanced Regression Modeling & Correlation Dynamics Project (2022-01 – 2022-06)
- Developed & Evaluated multiple linear and logistic regression models in R and Python.
- Diagnosed multicollinearity, heteroscedasticity, and outliers using VIF and Cook's Distance.
- Executed exhaustive correlation analyses (Pearson, Spearman, and Partial Correlation).
- Visualized correlation matrices and diagnostic heat maps for feature selection.
