Data Analyst

Analysis That Drives Decisions

Analyzed 1.2M+ real CFPB mortgage records. Delivered findings as interactive dashboards, Power BI reports (14 DAX measures), Excel workbooks, SQL case studies, and written business memos for non-technical stakeholders. 7 certifications, 917 hours of certified DA training.

Email Me Directly LinkedIn Download Resume
1.2M+ Rows Analyzed 25-Page Dashboard 7 DA/BI Certifications 11 SQL Case Studies Power BI (14 DAX Measures) Tableau dbt + Airflow Pipeline Evidence.dev BI Report

What the Analysis Found

Finance Analytics Portfolio — 1.2M+ real 2022 CFPB HMDA records

Credit Risk

DTI ratio, not FICO score, is the primary driver of loan default among prime borrowers (720+ FICO). Tightening DTI thresholds reduces expected losses by $600K-$900K per $100M portfolio with no impact on approval volume.

Fair Lending

County-level HMDA analysis surfaces geographic loan approval disparities with estimated $5-15M regulatory penalty exposure. Findings presented as a written memo with actionable compliance recommendations.

Loss Concentration

Grade B, not Grade G, is the true loss driver -- a $35M stressed-loss gap per $500M portfolio that conventional risk-tier framing misses entirely.

Tools & Pipeline

Full analytics stack — from raw data to boardroom-ready deliverables

Data Pipeline
  • SQL — PostgreSQL, DuckDB, Snowflake
  • dbt — 8 models, 33 schema tests, dual-target (DuckDB / Snowflake)
  • Airflow — 5-step TaskFlow DAG (ETL → GX → dbt → test → export)
  • Great Expectations — 3 validation suites, 26 expectations
BI & Visualization
  • Tableau — published dashboards (Public)
  • Power BI — 14 DAX measures, 3 report pages (PBIP)
  • Evidence.dev — 6-page BI report deployed on Vercel
  • LookML — 3 views, 3 explores, 30+ dimensions
Analysis & Python
  • Python — pandas, NumPy, scikit-learn, matplotlib, seaborn
  • Streamlit — 25-page interactive dashboard
  • A/B Testing — power analysis, z-test, Bonferroni, Bayesian
  • ML — Random Forest, K-Means clustering, SHAP
Reporting & Delivery
  • Excel — Pivot Tables, What-If, embedded charts, data validation
  • Google Sheets — collaborative reporting
  • Business Memos — findings framed as decisions with dollar amounts
  • SQL Case Studies — 11 standalone DuckDB query sets

Analytics Portfolio

Finance Analytics · Flagship

Finance Analytics Portfolio

Credit risk, market risk, and fair lending analysis on 1.2M+ 2022 CFPB HMDA mortgage records.

Deliverables

  • 25-page interactive Streamlit dashboard
  • 8-tab Excel workbook (pivot tables, What-If, charts)
  • Power BI report with 14 DAX measures
  • Evidence.dev 6-page BI report (deployed on Vercel)
  • 11 standalone SQL case studies
  • Written business memos with dollar-denominated recommendations

Tools

  • SQL (PostgreSQL, DuckDB, Snowflake)
  • dbt (8 models, 33 tests) · Airflow (5-step DAG)
  • Great Expectations (3 suites, 26 expectations)
  • Python (pandas, NumPy, scikit-learn)
  • Excel (Pivot Tables, VLOOKUP, What-If)
  • Power BI (DAX) · Tableau · Evidence.dev · LookML
Live Dashboard → GitHub → Tableau → Evidence.dev Report → Power BI DAX →
E-Commerce Analytics · Customer Retention

Olist E-Commerce Portfolio

Customer retention and delivery performance analysis on 100K+ real orders from a Brazilian e-commerce marketplace. Business question: what drives repeat purchase behavior and delivery satisfaction?

Key Findings

  • 93% of customers never return (vs 25-30% industry benchmark)
  • Late deliveries cost 1.2 review points on average
  • Top 5 high-delay states identified with carrier recommendations
  • K-Means: 4 actionable segments (Champions, Loyal, At-Risk, Hibernating)

Deliverables

  • 10 SQL queries (revenue, cohort, RFM, seller, delivery)
  • EDA + segmentation notebooks
  • Interactive dashboard
  • Written business memo (To/From/Date/Subject format)
Live Dashboard → GitHub →
Client Analytics · Real Stakeholder

Lyrio Real Estate Dashboard

Operational analytics dashboard delivered to a real estate firm. Replaced manual spreadsheet reporting with live KPI tracking reviewed weekly by operations management.

Lead conversion rates · Campaign costs & ROI · Pipeline KPIs
Live Dashboard →

DA/BI Certifications

Google Advanced Data Analytics 200 hours
Google Data Analytics 181 hours
IBM Business Intelligence Analyst 141 hours
Microsoft AI-Enhanced Data Analysis 120 hours
Microsoft Generative AI for Data Analysis 108 hours
Microsoft Data Visualization with Power BI 87 hours
Google Business Intelligence 80 hours

7 certifications · 917 hours · Google, IBM, Microsoft

Open to DA Roles — Remote

Targeting 50-500 employee companies where "data analyst" means the data function. Finance, operations, and business analytics preferred. Open to: Data Analyst, BI Analyst, Analytics Associate, Reporting Analyst, Operations Analyst.

US-based (Cathedral City, CA) · Canadian citizen, no sponsorship required · $30-45/hr or $50-75K salaried

caymanroden@gmail.com