Festus Udealor

hello_world.R · Bologna, Italy

Festus Onyeka Udealor

Data Analyst

I turn messy data into decisions. MSc in Statistical Sciences (Data Science) at the University of Bologna, first-class statistics graduate, and an analyst who works across R, Python, SQL and Power BI, from the query to the dashboard to the story.

Festus Udealor in a navy suit, standing outdoors
● open to roles · Dec 2026
4.84/5.00BSc CGPA, first class
3,502daily returns in my MSc thesis
B = 999bootstrap replicates for model choice
~500people at my UNIBO keynote
Festus standing at a conference table in a university meeting room

whoami

A statistician who likes to build things people can use.

I was born in Zaki-Biam, Nigeria, studied statistics at Federal University Wukari, where I graduated as the best student in my department, and moved to Italy for my master's at the University of Bologna.

My research is on robust clustering: finding the structure in financial returns while letting the extreme days, the COVID crash, the rate hikes, the tariff shocks, speak for themselves. Outside research I build dashboards and SQL pipelines, and I care a lot about explaining results clearly. I was named best presenter in my MSc cohort project.

Before data, I ran sales and inventory for my family's fashion business, so I think about customers and margins, not only models.

R (primary)PythonSQL / PostgreSQLPower BITableauEnglish C2Italian A2

ls ./skills

What I work with

The tools I reach for, from the first query to the last slide.

Programming

Radvanced
Pythonadvanced
SQL / PostgreSQLadvanced
Git / GitHubintermediate

Statistics & ML

Clustering & mixture modelsresearch
Regression & classificationadvanced
Time series (ARIMA)advanced
Bootstrap & robust methodsadvanced

Visualisation & BI

Power BIadvanced
Tableauadvanced
ggplot2advanced
Exceladvanced

Risk & finance

Financial return analysisresearch
Volatility & stress episodesresearch
Anomaly detectionadvanced
SPSSintermediate

git log --projects

Selected work

Research, dashboards and SQL analysis. Filter by the kind of work you care about.

RStudio on a laptop showing DBSCAN cluster assignments across financial return series from 2019 to 2026

MSc thesis · 2024–Dec 2026

Robust cluster analysis of financial returns

I compare GMM, TCLUST, OTRIMLE and DBSCAN on 3,502 daily log returns for Apple, Microsoft, Google, Meta, Amazon and the S&P 500, choosing models with a sequential parametric bootstrap (B = 999). The outliers line up with the VIX around COVID-19, the Fed hikes and the 2025 tariff shocks.

The methods agree on the extreme days but not on the structure inside the clusters, and that agreement weakens at k = 4.

Rtclustotrimledbscanmclustbootstrap

Supervised by Prof. Christian Hennig and Dr. Laura Anderlucci

Code, figures and data on GitHub ↗
dashboards/sales-overviewlive

Power BI-style dashboard · 2026

Festus Electronics sales BI

A two-page sales report built on the Contoso database: SQL view in PostgreSQL, then KPIs, trends, category mix and a country scorecard with click-to-filter.

$206M revenue across 83K orders; computers drive 44% of it.

PostgreSQLDBeaverdashboard
Open the two-page dashboard → Open the one-page version →
dashboards/toman-bikelive

Power BI · guided project by Absent Data

Toman Bike Shop dashboard

Excel to SQL to Power BI on a bike-sharing business: hourly profitability heat table, seasonal revenue and rider demographics. Rebuilt here as an interactive web dashboard over 34,758 hourly records from 2021 and 2022.

$15.2M revenue, $10.4M profit (69% margin), 81% registered riders.

ExcelSQLPower BIDAX
Open the dashboard →
churn / misclassification_error

Team project · Group 9 · UNIBO · 2025

Customer churn prediction

Churn model for 7,043 telecom customers, joined with their written complaints. We scored each customer's complaints with lexicon sentiment analysis, explored topics with latent semantic analysis, and compared logistic regression, lasso, boosting and random forest.

Complaint sentiment was one of the strongest churn signals (z = −10.5), next to tenure and contract type.

RglmnetgbmrandomForestNLP
SQL_Project_1 / top_paying_skills

SQL · 2024

Data job market analysis

Large-scale analysis of data job postings: which skills pay and which are in demand, using CTEs, window functions and ranking.

PostgreSQLVS CodeGitHub
View on GitHub ↗
Int_SQL_Project / cohort_analysis

SQL · 2024

E-commerce sales & retention

Customer segmentation, cohort analysis and churn modelling for an online store.

About 25% of customers generate about 66% of revenue.

PostgreSQLcohortsLTV
View on GitHub ↗
cad_ngn / arima_forecast.R
CAD to NGN monthly exchange rate, March 2010 to December 2020, rising from 95 to 299 naira, with an ARIMA(0,1,1) forecast of 302 naira through 2022 and a 95% interval widening to 231–463. CAD → NGN · monthly, 2010–2020ARIMA(0,1,1) forecast to Dec 2022 · 95% interval 100200300400201220142016201820202022 ₦302 2016 float
read the paper ↗

Publication · 2022

Forecasting the CAD/NGN exchange rate

ARIMA modelling with stationarity tests, diagnostics and a two-year out-of-sample forecast in R. My BSc project, completed in 2021 and published as Udealor & Adubisi (2022) on ResearchGate.

RARIMAforecasting
Read the paper on ResearchGate ↗
Lee-Carter-Italy-France / e0_forecast

Biodemography · UNIBO · 2025/26

Lee-Carter mortality: Italy vs France

Lee-Carter models fitted by SVD and Poisson MLE on Human Mortality Database data from 1872 to 2023. Includes a 20-year forecast, an Arriaga decomposition of the gender gap, smoking and health-spending context, and a Renshaw-Haberman cohort check.

The gender gap is forecast to plateau: 4.0 years in Italy and 5.5 in France by 2043. France's higher male smoking explains its larger gap better than health spending does.

RdemographyStMoMoforecasting
View on GitHub ↗
VS Code on a laptop with SQL scripts loading job-posting CSV files into PostgreSQL

behind the scenes

Where the dashboards start

Most of my work begins in a SQL editor: creating tables, loading raw CSVs, fixing encodings and writing the views a dashboard reads from. I'm working through advanced SQL and Python now, with window functions, CTEs and LTV analysis on the Contoso database.

cat experience.log

Experience & education

2024 – Dec 2026

MSc Statistical Sciences, Data Science track

University of Bologna

Statistical Learning, Advanced Time Series, Bayesian Inference, Big Data Analytics. Thesis on robust clustering of financial returns.

2025

Team Lead & Data Analyst, customer churn prediction

Data Science Applications, University of Bologna

Led modelling and the final presentation; recognised as best presenter.

Nov 2022 – Mar 2023

SPSS Tutor

PLASCHEMA, Plateau State Health Agency, Nigeria

Trained public health staff in data entry, cleaning, visualisation and interpretation, using epidemiological datasets.

Oct 2021 – Apr 2022

Data Systems Support, ICT

University of Jos, Nigeria

Maintained university data systems and fixed data quality issues to improve reporting.

2018 – 2022

Sales Manager, fashion retail

Family business, Nigeria

Ran sales, inventory and customer relationships, and improved customer retention.

2016 – 2021

BSc Statistics, First Class Honours

Federal University Wukari, Nigeria

CGPA 4.84/5.00. Best Graduating Student in Mathematics & Statistics; Academic Excellence Award (2020).

beyond the data

Community & speaking

Founder · 2023 →

De_Light Foundation

A non-profit in Nigeria giving young people tech-skills training and scholarship support, with the goal of raising 100 young techpreneurs by 2040.

Speaker

UNIBO International Student Day

Addressed about 500 attendees and was recognised for clarity and presence.

Volunteer

UNIBO Student Ambassador

Helps incoming international students settle in, from orientation to everyday life in Bologna.

Festus at a professional event beside a large screen showing a data presentation

./contact --now

Let's work with data together.

I'm looking for roles as a data analyst, statistician or quantitative researcher, as an internship or fixed-term position, from December 2026. I'm based in Bologna and happy to talk about remote work too.

Emailudealorfestus@gmail.com
Phone+39 351 685 0897
LocationBologna, Italy
LanguagesEnglish C2 · Italian A2