Lightnews — Scholar-powered news

Norm Matloff (你有冇諗清楚呀?)

@matloff.bsky.social

570 followers 910 following 1.5K posts

Em. Prof., UC Davis. Many awards, incl. book, teaching, public service. Many books, latest The Art of Machine Learning (uses qeML pkg). Former Editor in Chief, the R Journal. Views mine. heather.cs.ucdavis.edu/matloff.html

Posts Media Videos Starter Packs

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 7h

Elegant code is often more difficult to write, with subtler bugs, and difficult to debug. In the case of R, it might also run faster, but there is a tradeoff here, no consistent winner in all cases.

Reposted by Norm Matloff (你有冇諗清楚呀?)

ᴅʀ. ᴍᴏᴡɪɴᴄᴋᴇʟ'ꜱ @drmowinckels.io · 6d

Just published my new R article: 'Mapply: When You Need to Iterate Over Multiple Inputs'! 🚀 If `sapply` doesn't quite cut it for your multi-variable iterations, `mapply` is your friend. Learn to pair inputs beautifully. #RStats #Mapply
https://drmo.site/bhXeDb

A graphic showing the concept of mapply in R, with multiple input vectors being paired and processed by a function returning a single output vector.

8 29

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 1d

This really is an excellent book.

Big Book of R @bigbookofr.com · 1d

Modern Statistics for Modern Biology by Susan Holmes and Wolfgang Huber
#RStats
https://bigbookofr.com/chapters/life%20sciences.html#modern-statistics-for-modern-biology

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 2d

After years of hearing it, I still don't understand why stepwise is considered so evil by some.

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 2d

IMO, it's absurd to try to interpret a 4-way interaction. Even explaining 3-ways is questionable.

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 5d

Thanks, looks like great stuff on GMM, a topic I've always wanted to pursue but (like many others) have never had the time to do much.

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 5d

Thanks, nice work, looking forward to the preprint.

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 6d

I strongly agree. The function gtools::defmacro is very important in my work. (I'll give examples here is anyone is interested.) If you have some short misc. task to do, check gtools before writing your own.

Reposted by Norm Matloff (你有冇諗清楚呀?)

Big Book of R @bigbookofr.com · 6d

Time Series Analysis and Its Applications by Robert H. Shumway and David S. Stoffer
#RStats
https://bigbookofr.com/chapters/time%20series%20analysis%20and%20forecasting.html#time-series-analysis-and-its-applications

2 1

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 8d

Nice! But what is the theory behind it?

1 1

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 9d

Very nice!

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 11d

I certainly never noticed it. Good catch!

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 11d

What is the value of epsilon here?

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 12d

Excellent app!

Roger Day likes Golden Rule @knotcknot.bsky.social · 14d

Dr. Who solves the
"should we lump or should we split"
dilemma.
Compare lots of stat methods in this interactive app.
#rshiny #rstats #statistics #rstudio #bayesian #medical #treatment
Lumping&splitting.
Bias&variance.
Individuals&groups.
Borrowing strength
🧪
trials.shinyapps.io/Bias-varianc...

Lumping and splitting.
Bias and variance.
Individuals and groups.
“Borrowing strength”

Package:T15lumpsplit 1.6 2025-09-24
Author: Roger Day
An interactive “shiny” app.

1 2

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 15d

All should read.

But I object to the notion that suddenly AI is doing screening of job applications. As statisticians, we know this capability has existed for many, many years. They may have used a simple linear or logistic model, but it's really the same thing, just not as far-reaching.

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 15d

Link?

Reposted by Norm Matloff (你有冇諗清楚呀?)

Jennifer Davis @jmd5120.bsky.social · 16d

Hey

Bike lights

It's fall now

Chicago skyline at night from Shedd Aquarium

1 5 46

Reposted by Norm Matloff (你有冇諗清楚呀?)

David Blanchflower BSc @davidbflower.bsky.social · 16d

Uranus and the Pleiades. 22 September 2025. 🔭 🧪 🎨 #astrophotography #SciArt #photography #StormHour #ThePhotoHour

The Pleiades star cluster above centre. Planet Uranus dot with text label near bottom left corner. Many star dots.

6 28 180

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 16d

I'm told a highly-qualified Silicon Valley analyst was fired because he recommended using just linear regression in a very simple project.

1 2

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 18d

I prefer to write the proper equations, tinyurl.com/25nd5shd. But the key point is that writing equations or drawing a tree forces one to write down one's reasoning. The only reason MH is considered "hard" is that people try to intuit the answer rather than following the rules.

tinyurl.com

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 20d

All academics should read this!

Reposted by Norm Matloff (你有冇諗清楚呀?)

Daniel Gutierrez @ddgutierrez.bsky.social · 20d

I tell students: You don’t need to memorize every ML algorithm. But you do need to understand:
— loss functions
— overfitting
— regularization
Master the “why,” not just the “how.”

#DataScience #MachineLearning #Statistics #AI #RStats

1 7

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 21d

Yes! Thanks very much.

Reposted by Norm Matloff (你有冇諗清楚呀?)

Daniel Gutierrez @ddgutierrez.bsky.social · 21d

Most companies don’t need deep learning. They need clean joins, honest metrics, and interpretable results.
Startups often chase GPU clusters before fixing their CSVs.

#DataScience #MachineLearning #AI #RStats

6 14

Norm Matloff (你有冇諗清楚呀?) @matloff.bsky.social · 21d

Alas, in most courses the instructor is more excited about the subject matter than the students. My intellectual passions are in just two areas, coding and probabilistic modeling, and all the courses I taught were in these two categories. :-)