Nick Chaiyachakorn

chaiyach@ohsu.edu (academic CV tbd)
MS Biostatistics candidate
Oregon Health and Science University, School of Public Health

Path diagram of SEM latent growth model
Figure: path diagram of an SEM latent growth model, for 4 time points, with subject/case-varying constant and linear coefficient.

A bit about me. I am a quantitative researcher in the applied social and health sciences. I currently have the privilege of doing autism research in both community-participatory and epidemiological contexts. For fun, I take too many quantitative methods courses. (My graduate program directors can testify to the headaches I've given them!)

I aspire to be a statistical polyglot. So I've spent a lot of my career building a wide statistical toolbox: I've had the privilege to work with/learn from from social psychologists, survey statisticians, psychometricians, high-performance computing engineers, geostatisticians, and epidemiologists.

I have three core beliefs:

My Statistical Passport

As people collect visas to countries, I collect (and cherish) the diversity of statistical analyses I've done.

I've had the privilege of talking my research mentors into letting me doing the following types of analyses:


Structural equation modelling – exploratory/confirmatory factor analysis, multi-group SEM, longitudinal models, growth curve analysis
...on healthcare patient survey datasets
Multilevel/hierarchical/mixed effects modelling ...on high-school/post-high school educational attainment data
Analysis of survey sample data (e.g. multi-stage stratified/clustered samples) ...on public sector survey research at

...on epidemiological surveillance data (e.g. National Survey of Children's Health)
Latent variable methods in general – e.g. latent class and trajectory/transition analysis ...on healthcare accommodations data
Psychometrics and psychological scale/instrument validation ...on patient-reported outcome measures developed by and for autistic community members

These are things I've learned in graduate-level classes, and wish I could use more in real life...

Longitudinal models with separate cross-sectional and longitudinal effects (ahoy there Lord's Paradox!)
Matching methods for causal analyses (the types epidemiologists do)
Spatiotemporal modelling (co/variogram analysis and kriging, generalized additive models) ...on spatiotemporal ad auction data
Survival/time-to-event analysis (proportional hazards models, mostly)
 ...on breast cancer cohort study data

One day, I'll learn...

Complex survey weighting methods (raking)
 ...like the pollsters do!
Advanced design of experiments – the type that industrial statisticians use
 ...like the local statisticians at Intel do!

(back)

Writing

I learn things by writing them up. Here are some of the things I've been learning recently.

Caveat emptor: I'm not good enough at math to do derivations from scratch, so heavily rely on ChatGPT to give me rigorous mathematical derivations and arguments. For full transparency, I attach the prompts I used at the end.

(2025-12-01; unfortunately paywalled) Me and a research mentor, Dr. Katharine Zuckerman, co-authored a mini "op-ed" in the newsletter of the American Academy of Pediatrics about how autistic kids in crisis can be treated better by pediatric urgent care providers and hospitals.

A great deal, if not all, of these is true for autistic adults, and adults with disabilities in general.

Pediatricians are awesome. This benefitted heavily from the experiences of Dr. Zuckerman and her pediatrician colleagues who have thought a great deal about trauma-informed care of kids with autism (and other disabilities); especially those who experience crisis a great deal.

(2025-07-14) "Lab notebook": (Unsuccessful) attempt to convert graphics to ASCII art by subdivision into tiles and glyph matching.

(2025-04-02) Reminder to myself about the geometric intuition behind the OLS estimator + associated standard error.

(back)

Open-Source Software on GitHub

I've been getting back into coding; I do small projects that meet my needs/scratch my itches in my R statistical analyses, and vintage Macintosh hackery. Currently, I only have:

(2025-12-17) run_correlations.R Flexible R utility for running lots of correlations with optional stratification, directional/nondirectional hypothesis tests, and Bonferroni correction by default. Uses fancy tidyverse select syntax. Useful for psychometric validation studies.

color-detection-diagnostic.c Mini-utility exercising multiple methods to determine terminal colour capabilities on *nix platforms - environment variables, 'ncurses', xterm-style Primary/Secondary Device Attribute queries, true-colour roundtripping, a visual demo, and process name detection.

libfixmathmatrix A header-only (+ optional source file) amalgamation of a 32-bit fixed-point numerics and linear algebra library.

DazzlerGFX, heavily WIP/playground Mini vMac, modified to add a virtual "PDS expansion card" to the Macintosh SE platform. It currently has a general-purpose framework with (a) a whopping ~1 MB of shared on-device memory that can be accessed without any latency, and (b) the ability to asynchronously execute 7 commands at once. Just no actual functionality to emulate yet! You could probably even do DMA copies in one (emulated) cycle.

(back)

Research Mentors I've Been Privileged to Have

I'm indebted to so many kind people who've given advice to me, especially during rough times. Here, I'll limit myself to the academic mentors who I've had sustained relationships with for many years. They've kept me going rough times and the precarity of research. Alphabetical order:

(back)