◆ yatawara.com / teaching
Statistics and applied mathematics taught with computational thinking, real-world data, and student-centered active learning — where AI is integrated thoughtfully, only after the fundamentals are solid.
A first course in statistical reasoning — descriptive statistics, probability, sampling, estimation, and hypothesis testing — taught with real datasets and open educational resources so cost is never a barrier to learning.
The mathematical foundations of probability — random variables, distributions, expectation, moment-generating functions, and multivariate analysis — building the rigor behind every statistical method.
A general-education course using statistical tools — Lorenz curves, Gini coefficients, and disparity measures — to quantify and interrogate inequality in income, health, and the environment.
A graduate course in hands-on data analysis — programming statistical workflows and applying multivariate techniques to real problems using R and SAS.
The theory of statistical inference — point and interval estimation, properties of estimators, and the construction of hypothesis tests — derived and proven from first principles.
Building, diagnosing, and interpreting regression models — from simple linear regression through multiple, model selection, and diagnostics — with an emphasis on applied, real-world modeling.
Full interactive coursebook: lessons in English and Spanish, R and Python, chapter games.
Open the MATH 4210 coursebookA capstone for the Statistical Data Science concentration — statistical learning, classification, resampling, and model evaluation — taught from An Introduction to Statistical Learning with R and Python.
My teaching is rooted in a simple belief: education transforms lives. I learned this from my mother, who spent thirty-five years teaching in rural Sri Lankan schools. I watched her change the trajectory of students who had been told they had no future — and I carry that conviction into every classroom I walk into.
I think about teaching as four kinds of interaction, and I work to make each one strong:
I learn my students' names and stay approachable. Office hours are not a formality — they are where real learning often happens. In my classroom every question is equal; the one a student is afraid to ask is usually the one worth answering.
Statistics earns attention when it answers questions people actually care about. I motivate why it matters with real examples — how markets remember and forget, what crypto volatility reveals, how environmental exposure is distributed across communities — so the mathematics arrives with a reason to exist.
Students learn statistics by doing statistics together. Group projects and collaborative analyses turn a roomful of individuals into a working community of analysts who teach and challenge one another.
I push the work outward — into presentations, written reports, and conversations about careers — so that what students build in class becomes something they can carry into a job, a graduate program, or a research lab.
I use AI deliberately and ethically. My Custom GPT tutors come after the fundamentals, never as a substitute for them. The tools are there to deepen understanding, not to outsource it.
Anyone can google the steps to a hypothesis test, but you need to understand the basics to interpret the result and make inferences.
The job market has changed. My duty to my students is to prepare them for the one that exists now — equipping them with computational skills, AI literacy, data-science workflows, and the critical thinking to work alongside AI rather than be replaced by it.
$ tutors --load custom-gpt --gate fundamentals-first
A guided companion that helps elementary-statistics students learn R step by step, reinforcing concepts already covered in class.
A mathematical-statistics tutor that walks through derivations and inference carefully — supporting proof-based reasoning, not shortcutting it.
Turns handwritten or plain-text mathematics into clean LaTeX, lowering the friction of producing professional technical writing.
Language-specific coaches that help students build fluency across the computing environments used throughout the curriculum.
An approachable tutor for foundational mathematics, meeting earlier learners where they are and building confidence.
A study-strategy partner that helps students plan, practice, and prepare — building habits, not just answering questions.
| Program | Year | Students | Project |
|---|---|---|---|
| SURE (Chevron) | Summer 2025 | J. Rosas, J. Rodriguez, C. Rodriguez, R. Gamez | Local LLM applications for education |
| CV Pathway | Summer 2025 | E. DeJesus, N. Gallego | Air pollution & fertility patterns |
| Student Research Scholars | 2024–25 | 1 student | “How Present is ChatGPT at CSUB?” |
| CV Pathway | Summer 2024 | T. Regpala, J. Rodriguez Aguilar | Air pollution research |
Find my current office hours on your course Canvas page, or email me to set up a time. If you are stuck, come early — I would much rather help before an exam than after.
I am glad to write letters for students I know well. To write you a strong one, I need:
Email me anytime at ayatawara@csub.edu.