Lightnews — Scholar-powered news

Kert Viele @kertviele.bsky.social · 8d

It’s a really bizarre business model to me. It seems aimed at SEEMING like our standard data analysis paradigms, but it’s much like me deciding to fit a linear regression and instead of telling you the coefficients, I instead give you a bunch of (X,Y) pairs generated from my fitted model.

2

Kert Viele @kertviele.bsky.social · 8d

None of this is properly “data”. At best it is a snapshot of a predictive distribution from some black box AI model fit. What you are estimating from such things is, at best, the AI model fit from the actual data (which the AI could just provide directly instead of “data” but often doesn’t).

1 3

Kert Viele @kertviele.bsky.social · 14d

Feels a bit like dominantly inherited Alzheimer’s. Genetic basis, somewhat predictable age of onset, etc.

1 2

Kert Viele @kertviele.bsky.social · Sep 3

From a modeling perspective we can have a debate. But in reality sites are far from random. Sponsors have preexisting arrangements with productive sites etc.

2 5

Kert Viele @kertviele.bsky.social · Aug 29

Other applications of hierarchical models here for borrowing across different regions of the world.

cdn.who.int/media/docs/d...

cdn.who.int

1 1 1

Kert Viele @kertviele.bsky.social · Aug 29

After a month I can’t remember what my own code does by reading the source (well at least it takes a while, and I don’t think I’m the worst coder). My help files are for me. If they help someone else too, great.

1 1 5

Kert Viele @kertviele.bsky.social · Aug 28

It's unclear the conclusions in that paper...that essentially the extra complexity isn't worthwhile...apply outside of the context of that paper, particularly the small sample sizes.

1 1

Kert Viele @kertviele.bsky.social · Aug 28

I don't know that I have seen a writeup specific to historical borrowing (too many things like that...). However the models are essentially identical to those in the basket trial literature like this paper.

www.tandfonline.com/doi/abs/10.1...

A Comparison of Different Approaches to Bayesian Hierarchical Models in a Basket Trial to Evaluate the Benefits of Increasing Complexity

A basket trial is a trial that investigates a single experimental therapy simultaneously across multiple different patient population groups. When groups are heterogeneous, the optimal analysis is ...

www.tandfonline.com

2 1

Kert Viele @kertviele.bsky.social · Aug 26

There are multiple versions. Some are simply a hierarchical model over studies. Others fit a two component mixture with one component having the current study independent of others. Others fit more complex models like dirichlet processes.

1 2

Kert Viele @kertviele.bsky.social · Aug 19

Utilities. If you aren’t using good ones explicitly, you are likely using bad ones implicitly.

Adam L @lauretig.bsky.social · Jul 28

47. People really don't know what to do with ordinal variables, so they either treat them as continuous or dichotomize them

Kert Viele @kertviele.bsky.social · Aug 17

Of course, seeing people submit false AI references can be very informative about their quality control….we have already caught false references in potential FDA submissions.

1 1

Kert Viele @kertviele.bsky.social · Aug 15

I can see my kids adding "but we are still collecting data to increase power".....

Charnel Pleasures @charnel-pleasures.bsky.social · Aug 14

After like a decade and a half of free-riding on @zachweinersmith.bsky.social’s public goods, I finally decided to engage in a Coasean bargain to compensate for my consumption of his externalities.

@smbccomics.bsky.social #SMBC #Economics #stats

1

Kert Viele @kertviele.bsky.social · Aug 15

But...but....I really like my
Bayesian Adaptive Design And Subject Simulator
software....

1 11

Reposted by Kert Viele

Scott Ashworth @soashworth.bsky.social · Aug 15

Remember kids:

A regression of height on weight answers questions like:
Bob weighs 200 pounds. What's your best guess for his height?
It is useless for questions like:
If Bob went on a diet and lost 20 pounds, how much less would his height be?
The answer to that one is "not at all", no matter what any regression says

2 13

Kert Viele @kertviele.bsky.social · Aug 12

I think this issue is one of the most appealing aspects of Bayes. With Bayes the posterior distribution is the basis for inference, and it is invariant to the design.

Triad sou. @triadsou.bsky.social · Aug 11

Confidence Intervals for Adaptive Trial Designs II: Case Study and Practical Guidance. David S. Robertson, Thomas Burnett, Babak Choodari-Oskooei, Munya Dimairo, Michael Grayling, Philip Pallmann, Thomas Jaki. Statistics in Medicine. onlinelibrary.wiley.com/doi/10.1002/...

Confidence Intervals for Adaptive Trial Designs II: Case Study and Practical Guidance

In adaptive clinical trials, the conventional confidence interval (CI) for a treatment effect is prone to undesirable properties such as undercoverage and potential inconsistency with the final hypot...

onlinelibrary.wiley.com

2 1

Kert Viele @kertviele.bsky.social · Aug 11

And you thought reviewer 2 was rough…

Derek Beres @derekberes.bsky.social · Aug 11

A large-scale, government-funded study tracking 1.2M children over two decades found something Kennedy doesn't like, so he wants it retracted.

www.reuters.com/business/hea...

Exclusive: Medical journal rejects Kennedy's call for retraction of vaccine study

An influential U.S. medical journal is rejecting a call from Health Secretary Robert F. Kennedy Jr. to retract a large Danish study that found that aluminum ingredients in vaccines do not increase health risks for children, the journal's editor told Reuters.

www.reuters.com

Kert Viele @kertviele.bsky.social · Aug 6

Agree with proper terms in the model the hybrid isn’t itself bad, although if you can individually randomize I would strongly recommend doing that from the start. Separate issue if you think one part of the hybrid is biased.

1 1 2

Kert Viele @kertviele.bsky.social · Jul 31

I can’t comment on the percentage of the industry that does such due diligence as standard practice, but we have done a lot of consulting for venture capitalists to vet what they are potentially investing in.

1 2

Kert Viele @kertviele.bsky.social · Jul 22

Ah you got me. It’s not that old (maybe 5 years) but my beard has turned grey in the meantime! It’s a good picture because my son was joking around in the background.

1

Kert Viele @kertviele.bsky.social · Jul 22

The right frequency is based on how often you take a good picture. Those simply could be 20 years apart for some of us.

1 1

Kert Viele @kertviele.bsky.social · Jul 17

Agree lots of problems implementing this. Amusingly this may mark the return of the travel agent, suddenly back to manipulate the AI that is trying to manipulate you.

2

Kert Viele @kertviele.bsky.social · Jul 10

Said like a true Bayesian

1

Reposted by Kert Viele

JAMA @jama.com · Jul 9

📢 Apply now for the JAMA Network Peer Review Academy!

This 9-month remote program connects early-career researchers with JAMA Network editors for hands-on peer review training.

https://ja.ma/3TlIeIO

Build your peer review expertise. Applications open now.

2 3

Reposted by Kert Viele

Steven Tong @steventong.bsky.social · Jul 8

The Staphylococcus aureus Network Adaptive Platform (SNAP) trial has been nominated for a Patient healthcare communication award!

Public voting is now open 👉 ccrew.accesscr.com.au/cpir-2025/pu...

Find out more about SNAP 🔗 www.snaptrial.com.au

Give us a vote!!

@thedohertyinst.bsky.social

SNAP Trial – Staphylococcus aureus Network Adaptive Platform Trial

SNAP aims to improve treatment outcomes for patients with Staphylococcus aureus bloodstream infections. This website provides information for participants, members of the public, and healthcare profes...

www.snaptrial.com.au

3 18

Kert Viele @kertviele.bsky.social · Jul 7

“Would you consider an adaptive trial”

“No we’re going to enroll too fast”

“Are you sure? Some kind of adaptation will be a partial insurance policy against slow enrollment”

“We are sure”

One year later….

“Can we amend our protocol….”

1