Nick Chaiyachakorn

These were analyses I helped conduct in the first half of 2026...

• Working with longitudinal cohort data from the Health and Retirement Study, I:
  • ran survey-weighted survival analyses, to...
  • ...compare the predictive utility of measures of "frailty" wrt. to post-retirement lifespan.
• For a local health system's warehouse division, I:
  • consulted on methods of estimating demand for medical supplies...
  • ...based on multiple years of discrete transaction data.
• For an autism research group I am being mentored in, I:
  • conducted analyses of psychometric reliability and validity..
  • ...to psychometrically validate a measure of overall health created with and for autistic adults, using multi-year pilot cohort data.
• For a geographer collaborating with an autism research group I intern with, I:
  • linked Census data between multiple types of geographic units...
  • ...to identify areas with native communities impacted by tribal health infrastructure.

I learn things by writing them up. Unfortunately, since grad school is killing me, I don't have a lot of time to wring things up. Nevertheless, here are some of the things I've been learning recently.

Trauma-informed care principles can be adapted for youth with autism. AAP News. (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.

Geometric interpretation of the OLS estimator (2025-04-02)

Reminder to myself about the geometric intuition behind the OLS estimator as a projection operator, and its associated standard error.

Some R utility functions I've created in my time doing data analysis, that doesn't deserve a whole library. Other coding projects I've done when procrastinating from the actual data analyses I'm supposed to do.

R utility: save.flexibly - save with elements/names specified in a list. (2026-03-24)

R's save function is a little hard to use in data pipelines, since it requires the values to be saved to be actual assigned variables. Here, save.flexibly allows you to specify values to be saved as a list of the form list("name1", value1, "name2", value2, ...)

               
# `save.flexibly` -- frontend to R `save` that takes names and values of objects
#   to be saved as a list of the form list("name as string", value, "second name",
#   second value...)
# Arguments:
#  `save.flexibly(
#     list(name1, object1, name2, object2...),
#     ...arguments to 'save'...
#. )`
# 
save.flexibly <- function(names.vals.list, ...) {
  # Make sure 'names.vals.list' is a list of names
  # followed by objects:
  if(0 != (length(names.vals.list) %% 2))
    stop("save.flexibly: 'names.vals.list' doesn't have an even number of items (i.e. name-value pairs")
  should.be.names <- names.vals.list[seq(1, length(names.vals.list), 2)]
  `all.names?` <- Reduce(`&`, lapply(should.be.names, is.character))
  if(!(`all.names?`))
    stop("save.flexibly: an odd element of 'names.vals.list' (1, 3, ...) isn't a string giving an item name")
  
  # Okay, assign the elements...
  for(base.i in seq(1, length(names.vals.list), 2)) {
    assign(names.vals.list[[base.i]], names.vals.list[[base.i + 1]])
  }
  
  # And now save in one go; this requires synthesizing a call...
  # ...of the form save(...names..., ...other arguments...)
  to.call <- as.call(
    append(
      list(as.symbol("save")),
      append(
        should.be.names,
        as.symbol("...")
      )
    )
  )
  eval(to.call)
}
               
               

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.

Print alignment/printable bounds test page (letter-size)

Have a questionable printer? Want to figure out how exactly the image on your computer screen actually gets mapped to a physical page? Need to detect whether your pages print misaligned, or need to determine what parts of the printed page get cut off? Look no further!

Print this test page out at a 1:1 ratio. To determine alignment, fold the page horizontally and vertically, and measure where the creases cross the horizontal and vertical rulers. To determine where the printing bounds end (assuming that the page is being printed 1:1), look at where the horizontal and vertical rulers get cut off.

Exercise for the reader: correct for misalignment on your computer, without adjusting the printer.

Mirror: M1 Exploration - v0.70, Maynard Handley

Have you barely learned what a superscalar processor is? Have you always wondered how modern microprocessors actually keep track of instructions "in flight"? Here's an old document by Maynard Handley providing a surprisingly accessible nuts-and-bolts description of the microarchitecture of Apple's M1.

This was circulating on the internet around 2021 via a single Google Drive link, but disappeared a while ago. To preserve it for posterity, I mirror it here.

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:

• Debi Elliott (Portland State University, Regional Research Institute)
  
• Christina Nicolaidis (Portland State University, Academic Autism Spectrum Partnership in Research and Education)
• Katharine Cahn (emerita, Portland State University, Center for Improvement of Child and Family Services)
• Katharine Zuckerman (Oregon Health and Science University, Pediatrics)
• Lisa Steinman (emerita, Reed College)
• Mathew Uretsky (Portland State University, School of Social Work)