Statistics and Data Analysis — MBA

Applied statistics and machine learning coursework/thesis applying data-driven methods to environmental ecology questions.

Overview

An MBA in Data Science and Analytics (University of São Paulo), completed alongside my PhD, applying statistical and machine learning methods to microbial environmental ecology questions.

Scientific or practical problem

Ecological datasets are often high-dimensional and noisy; extracting valid, generalizable patterns requires rigorous statistical and machine learning practice — not just running models, but validating them appropriately.

My role

Student researcher — self-directed thesis and coursework.

Dataset or data type

Environmental/microbial ecology datasets from my PhD research, reanalyzed with a broader statistical/ML toolkit.

Methods and technologies

• Exploratory data analysis and data visualization
• Multivariate statistical analysis
• Regression and predictive modeling
• Machine learning for classification and feature selection
• Principal component analysis and dimensionality reduction
• Clustering and pattern recognition
• Model validation and performance evaluation
• R, Python, and reproducible analytical workflows

Workflow diagram

flowchart LR
  A[Raw dataset] --> B[EDA & visualization]
  B --> C[Feature selection / dimensionality reduction]
  C --> D[Model training: regression / classification / clustering]
  D --> E[Validation & performance evaluation]

Major results or outputs

Thesis: Unveiling microbial environmental ecology through data analysis techniques.

Challenges and decisions

Applying formal model-validation practice (cross-validation, held-out testing) to ecological data with limited sample sizes — a discipline that has carried into my subsequent bioinformatics work.

Publications

MBA thesis (not separately peer-reviewed).

Code and documentation

Not publicly released.

Collaborators

Independent coursework/thesis; supervised at University of São Paulo, Luiz de Queiroz College of Agriculture (ESALQ/USP).

Current status

Completed (2023).

Limitations or confidentiality note

None.