Md Iqbal Hossain

PhD Candidate  ·  Old Dominion University  ·  Norfolk, VA

Md Iqbal Hossain

Statistician specializing in Dependency Structure Analysis, Causal Dependency Analysis though DAGs, Copula Modeling, Count Time Series Data, and Tropical Cyclone Count Data Analytics: advancing computational statistics with applications in count time series, climate science, public health, and electoral behavior.

View My Research Get In Touch
Actively seeking Post-PhD opportunities starting Fall 2026

Actively seeking Post-PhD roles across academia and industry, focusing on advanced statistical modeling and data science.

Postdoc
Teaching
Data Scientist
Statistician
Biostatistician
Research Scientist
Statistical Analyst
4+
Research Articles
4
Conference Presentations
8+
Years of Teaching
4
Academic Degrees

Who I Am

Professional Statement

I am a Ph.D. Candidate in Computational and Applied Mathematics (Statistics) at Old Dominion University, with expertise spanning advanced statistical modeling, count data analysis, machine learning, and public health data science.

My doctoral research, supervised by Dr. Norou Diawara, centers on developing a novel copula-based multi-series dependency framework for modeling count time series (e.g., tropical cyclone counts): incorporating structural change-point detection and copula-based directed acyclic graphs (cDAGs) to map storm interdependence across the count time series (e.g., Atlantic, Pacific, and Indian Ocean basins using IBTrACS data for tropical cyclone counts).

I bring a proven record in teaching, statistical consulting, and cross-disciplinary collaboration, connecting rigorous methodology to real-world problems in climate science, electoral behavior, and clinical prediction.

Email Me Download CV

Statistical Modeling

Copulas, time series, count data, structural change-point analysis

Programming

R, Python, SAS — expert-level statistical computing

Public Health

Machine Learning and Bayesian Models for Early-Stage Diabetes Risk Prediction

Climate Analytics

Tropical cyclone modeling · IBTrACS global data

Teaching

8+ semesters instructing statistics at ODU & CMU

Competencies

Technical Skills

Programming Languages

R Python SAS MySQL Fortran

Statistical Methods

Copula Models Time Series Count Data Analysis Bayesian Inference Change-Point Analysis DAG GLM / Regression

Machine Learning

SVM Decision Trees Random Forest Logistic Regression PCA Neural Networks

Data Visualization

Tableau ggplot2 Matplotlib Seaborn Plotly

Statistical Software

SAS Enterprise Miner SPSS Minitab Stata

Certifications & Awards

  • SAS Certified Base Programmer for SAS 9
  • 2nd Place — VA-ASA Poster Competition (2023)
  • Graduate Teaching Assistant Development Program

Academic Work

Research & Projects

PhD Dissertation  ·  Ongoing

Copula-Based Multi-Series Dependency Framework for Count Time Series Data (Tropical Cyclone (Hurricane) Counts)

Advisor: Dr. Norou Diawara  ·  Old Dominion University

Developing a novel statistical framework for dispersed, over/under-dispersed time-series count data using the copula technique. The model integrates structural change-point detection and directed acyclic graphs (DAG) to map the interdependence across count time series (e.g., for tropical cyclone counts to map the interdependence in the Atlantic, Pacific, and Indian Ocean basins. Applied to decades of global hurricane data from the IBTrACS (International Best Track Archive for Climate Stewardship)).

Dataset
IBTrACS: Global, Decadal
Core Methods
Copulas · DAG · Change-Points
Data Type
Count Data · Time Series Data
Tropical Cyclone Spatial Dependence Network (1980-2024)
Master's Project · 2022

Early-Stage Diabetes Risk Prediction: Large-Scale ML & Bayesian Benchmarking

Supervisor: Dr. N. Rao Chaganty, ODU

A comprehensive benchmarking study comparing Logistic Regression, SVM, Decision Trees, and multiple ML/Bayesian models for early diabetes detection. Motivated by CDC data: 37M+ US adults with diabetes, 1 in 5 undiagnosed. Completed project extended to a published large-scale comparative analysis.

SVM Decision Trees Bayesian Models Logistic Regression Public Health
ML Algo Final
Research Assistant · 2020–2022

Voter Behavior, Turnout & Suppression Analysis of 2020 U.S. Election

ODU · School of Public Service & Dept. of Mathematics & Statistics

Analyzed the ANES 2020 and Harvard's Cooperative Election Study 2020 datasets to model the probability of individual voter participation and identify systemic factors driving voter suppression and turnout disparities in the 2020 U.S. Presidential Election. Applied logistic regression, random forest models. Presented the findings to interdisciplinary audiences in political science and statistics.

ANES 2020 CES 2020 Voter Suppression Turnout Modeling Logistic Regression
Voter Behavior
Undergraduate Project · 2016–2017

A Study on the Evaluation of Hospital Facilities in Bangladesh: Heart Patients

Supervisor: Mohammad Ahsan Uddin, Assoc. Prof., University of Dhaka

Designed survey instrument, collected and cleaned data, and performed a three-phase analysis: univariate (frequency tables, cross-tabulations), bivariate (chi-square tests, ANOVA), and multivariate (KANO model, Principal Component Analysis) to evaluate hospital quality from patients' perspectives.

Survey Design ANOVA PCA KANO Model Chi-Square

Scholarly Output

Publications & Presentations

Research Articles

  1. 1

    Applying Changepoint-Copula Modeling to Count Time Series Analysis: an Example of Tropical Cyclones

    This study introduces a framework combining changepoint detection and copula modeling to map shifting global tropical cyclone dependencies between 1980 and 2024. Results reveal a major regime shift in 2000, marked by a 59% surge in North Atlantic storm frequency and a move toward stronger, more complex extreme-value dependencies across basins.

    Count Time Series Dependency Network Structrual Break Dependency Structure
  2. 2

    Comprehensive Benchmarking of Several Machine Learning and Bayesian Models for Early-Stage Diabetes Risk Prediction: A Large-Scale Comparative Study

    Systematic comparison of predictive frameworks for clinical decision support in early diabetes detection, covering SVM, decision trees, random forests, and Bayesian approaches.

    Machine Learning Public Health Bayesian Statistics
  3. 3

    Contemporary Voter Suppression: Impact on the 2020 General Election

    Quantitative analysis of systemic barriers to voting and their measurable influence on election outcomes using large-scale electoral datasets.

    Political Science Applied Statistics
  4. 4

    In Search of the Rational Voter in the 2020 Presidential Election: Understanding the Impact of Voter Costs and Benefits on Turnout

    Empirical examination of rational choice theory applied to 2020 voter participation patterns, using probabilistic models on ANES and CES data.

    Electoral Behavior Rational Choice Theory

Conference Presentations

2026 Oral Presentation (25 mins)

ODU Mathematics Awareness Conference 2026

Old Dominion University, Norfolk, VA

2025 Poster Presentation

American Statistical Association — Virginia Chapter (VA-ASA)

Virginia Tech, Blacksburg, VA

2024 Oral Presentation (20 mins)

Midwest Political Science Association Annual Conference

Chicago, IL

2024 Poster Presentation

American Statistical Association — Virginia Chapter (VA-ASA)

University of Virginia, Charlottesville, VA

2023 2nd Place Award

American Statistical Association — Virginia Chapter (VA-ASA)

Virginia Commonwealth University, Richmond, VA

Career History

Professional Experience

Research Experience

2023 – Present

Graduate Research Assistant

Old Dominion University · Norfolk, VA

Developing copula-based multi-basin tropical cyclone count models with structural change-point analysis and DAG interdependence mapping. Applied to global IBTrACS hurricane data.

2020 – 2022

Graduate Research Assistant

ODU · School of Public Service & Dept. of Math & Statistics

Voter behavior and turnout analysis using ANES 2020 and Harvard CES 2020 data. Modeled voter suppression determinants in the 2020 U.S. Presidential Election.

Summer 2019

Statistical Consulting Intern

Central Michigan University · Statistical Consulting Center

Under Dr. Chin-I Cheng. Tableau training & dashboard design; carotid artery stenosis imaging study (US/CTA/ICA); flu vaccine response analysis by zip code; MAT survey analysis (student & non-student cohorts).

Teaching Experience

Old Dominion University

Norfolk, VA

2021 – 2026
  • STAT-130M · Elementary Statistics — Instructor (Fall 2021, Fall 2022, Spring 2023, Fall 2024)
  • STAT-603 · Probability Models for Data Science — Teaching Assistant (2024–2026)
  • STAT-604 · Statistical Tools for Data Science — Teaching Assistant (2024–2026)

Central Michigan University

Mt. Pleasant, MI

2018 – 2020
  • STA-282QR · Introduction to Statistics — Instructor (Fall 2019, Spring 2020)
  • MTH-105 · Intermediate Algebra — Instructor (Fall 2018 ×2 sections, Spring 2019)

Academic Background

Education

In Progress · Expected Summer 2026

Ph.D. Candidate

Computational & Applied Mathematics (Statistics)

Old Dominion University · Norfolk, VA

Dissertation: Dependency Structure and Causality Analysis though DAGs and Copula Modeling with Change Point Detection for Count Time Series Data

2020 – 2023

Master of Science

Computational & Applied Mathematics (Statistics)

Old Dominion University · Norfolk, VA

2018 – 2020

Master of Science

Applied Statistics and Analytics

Central Michigan University · Mt. Pleasant, MI

2013 – 2017

Bachelor of Science

Statistics

University of Dhaka · Dhaka, Bangladesh

Let's Connect

Get In Touch

Currently on the academic and industry job market. Open to research collaborations, faculty positions, data science roles, and consulting opportunities.

ODU Email
mhoss005@odu.edu
Personal
iqbal.kmbd@gmail.com
GitHub
@iqbal1012
LinkedIn
https://www.linkedin.com/in/md-iqbal-hossain-stat

Send a Message