Frame the question
Define the cohort, outcome, units, and observation window before selecting a method.
BIOSTATISTICIAN · RESEARCH ANALYST
Research data.
Statistical depth.
Clear answers.
I build statistical workflows that make research questions measurable, results reviewable, and uncertainty visible.
Aviation sleep, fatigue, and cognitive-performance research. Public summaries only; participant-level data stays private.
FROM OBSERVATION TO EVIDENCE
01 / FRAMEDefine the outcomeBaseline · eligibility · time
02 / ANALYZETest the modelAssumptions · diagnostics
03 / EXPLAINShow the limitsEvidence · uncertainty
Illustrative workflow, not participant data
SCROLL TO THE CASE STUDIES
4routes in the recovery analysis
3,198flight records in a saved validation check
24 / 48 / 72hour sleep-feature windows
Project-specific scope from reviewed materials; figures are not population-wide outcomes.
01 / SELECTED WORK
Seven ways I work through complexity.
Each case study goes beyond a tool list: the question, the implementation choices, the checks, and what the evidence can actually support.
7 projects
MODELING / 01VIEW CASE STUDY
01 / 07 Four route analysis scope
Cognitive recovery after long-haul flight
How do you define and model a return to baseline when some recoveries are never observed?
Explore projectDATA ENGINEERING / 02
VIEW CASE STUDY
02 / 07 24 / 48 / 72 hour windows
Sleep data integration across trips
Making wearable exports, sleep logs, and flight records agree on who, when, and which trip.
Explore projectRESEARCH TOOLS / 03VIEW CASE STUDY
03 / 07 Routes / crew / phases
Interactive aviation research dashboards
Turning complex research cohorts into explorable views without hiding denominators.
Explore projectRESEARCH TOOLS / 04
RULESCHECKREVIEW
VIEW CASE STUDY04 / 07 3,198 records checked
Flight-duty screening and output validation
A traceable rules workflow with a separate check of the saved outputs.
Explore projectSTUDY DESIGN / 05VIEW CASE STUDY
05 / 07 10,000-replicate checks
Study power and repeated-measures simulation
Testing recruitment assumptions before a study is conducted.
Explore projectMODELING / 06
VIEW CASE STUDY
06 / 07 Four route analysis scope
Inflight rest and pre-flight nap analysis
Comparing rest patterns while keeping subgroup size and uncertainty visible.
Explore projectDATA ENGINEERING / 07
VIEW CASE STUDY
07 / 07 Route-by-month summaries
Pilot metrics and reporting automation
Keeping monthly research summaries consistent as records and program rules evolve.
Explore projectResearch stays responsible.
These pages explain methods and capabilities without publishing university datasets, participant identifiers, internal workbooks, or unverified final findings.
FEATURED / VALIDATION
Trust grows when the output can be checked.
The flight-duty screening project is a useful example: rules are explicit, outputs retain an audit trail, and a separate validator compares saved results with source records.
See the validation case study01Normalizekeys, time zones, segments
02Screenconfigured rule branches
03Reconcileaudit fields and duplicates
3,198 records checked in saved summaries
02 / THE APPROACH
Rigorous, from the first join to the last sentence.
Biostatistics is a chain of decisions. I keep the chain visible so collaborators can challenge it, reproduce it, and use it.
Make data comparable
Align records and timestamps, document exclusions, and preserve denominator definitions.
Test the analysis
Choose models for the study design; inspect assumptions, missingness, and sensitive cases.
Explain the boundary
Share what the evidence shows, what remains uncertain, and where a conclusion should stop.
03 / ABOUT
Statistics with a reason to be understood.
I am a biostatistician whose work spans longitudinal research data, survival and mixed-effects modeling, interactive R tools, and reproducible reporting. My background also includes graduate research in statistical science and mathematics.
I care about the practical middle of research: clear definitions, trustworthy transformations, checks that catch mistakes, and communication that helps a multidisciplinary team act on the right conclusion.
CAPABILITIES
Statistical methods
Cox regression, survival analysis, mixed models, repeated measures, simulation, diagnostics
Data & software
R, tidyverse, lubridate, Shiny, Plotly, DT, openxlsx; Python and SQL
Research communication
R Markdown, Quarto, ggplot2, cohort summaries, Excel reports, stakeholder-ready explanations
04 / CONTACT
Have a research question worth making clearer?
I’m happy to discuss methods, data workflows, or how the work in these case studies could translate to your team.