#bayesian statistics

[gynoidgearhead]

[Image caption for above addition: two comics, each headlined "Mental Gymnastics". The first is a woman or girl stepping a few steps on a gymnastic mat and then happily extending her arms, with a caption reading "I guess fairies are real."

The second has a man energetically swinging from a bar, first from his hands and then his feet; then swinging from rings; then using a pommel horse; then jumping over a flaming car while in a superhero outfit. The various steps are captioned: "A walrus has escaped a zoo or decided to leave its territory for some reason then traveled hundreds of miles of waterways and rivers without being noticed." / It then decided to leave the water and start traveling on land specifically in my city, traveling across miles of streets without being run into or stopped." / "Then it came to my house in particular, made its way to the door, then knocked on it and politely waited." End caption.]

Or someone decided to pull a very illegal prank on you and deliver a walrus to your doorstep, in the manner of xkcd's "instead of office chair, package contained bobcat. would not buy again".

I've asked this question before and been surprised by the results, now I have access to more weirdos it's your problem:

It is the middle of a Sunday afternoon. You have nothing on, and aren't expecting visitors, deliveries or post.

Unexpectedly, there is a knock at the door.

[charliefooks]

Literally everything about this

[imkeepinit]

Bayesian statistics provides us with mathematical tools to rationally update our subjective beliefs in light of new data or evidence.

[soph-likes-space]

Tag urself im a nuisance parameter

[silelda]

A new addition to my Favorite Acknowledgements:

To the exclamation point. We couldn't talk about our friends, family, or Bayesian statistics without you.

- Bayes Rules! By Alicia A. Johnson, Miles Q. Ott and Mine Dogucu

[stabby-wabby]

My feelings about studying Bayesian Statistics

[mirastudiesphysics]

Meant to come back to this, but research got in the way and I got a week long extension.

By the third problem I realized that I could make my computer do most of the calculations for me in python, so I ended up programming a simple way to calculate each of the marginals (the probability density of rolling a 6 on a die from 0-5 for 5 rolls) for a binomial distribution, as well as the posteriors for the parameters. The parameter in this problem describes the probability that the die is weighted towards or against rolling a 6. This is done a few times, since the posterior you calculate can be used as a new prior for the next set of data.

Really need to work on my bayesian statistics hw so I'm posting here for once for accountability. I'd like at minimum to reblog with something I learn a long the way.

[art-of-mathematics]

Title: "Non-linear regression in a blurry cloud of (un-) certainty"

Date: 2023/06/12 - Size: DIN A4

Collage made with torn pieces of paper, printed background paper (top is rather dark night sky, bottom is mererly clouds in pastel-colors)

I resized and printed the non-linear regression visualisation/illustration and put it on top of the watercolour background paper.

I included a scrap piece of paper with the title of the picture and have torn it with a spiral-shaped jag at the bottom, which I bent around the top part of the non-linear regression illustration.

[psycheapuleius]

[gammaraydeath]

Phenetics :) cladistics :) statistical phylogenetics 😱

[shakespearenews]

Sam Bankman-Fried is wrong about Shakespeare

This is an ostensibly logical argument based on quantitative information. It doesn’t rely on subjective judgements: it just claims that it’s unlikely that Shakespeare is the GOAT because the population of literate individuals in his time was tiny compared to later centuries. On numbers alone, we’re more likely to find the greatest ever writer among the educated multitudes of the 20th and 21st centuries, not the thinly spread turnip-munchers of the 16th.

[joshuapaulbarnard]

Predicting Wine Quality

Predicting Wine Quality by comparing Linear Regression with Machine Learning techniques. Comparing Linear Regression with kNN, Decision Tree and Random Forest with Bayesian Inference to Predict Wine Quality in Python. We use python and Jupyter Notebook to download, extract, transform and analyze data about the physicochemical properties which make up wine, and use them to predict…

[rachelraygifs]

I’m a frequentist* not a bayesian so maybe I’m already biased but literally anyone who talks about their “priors” in real life should be shot out of a canon

[soph-likes-space]

She prior on my likelihood till i posterior

[dorianbrightmusic]

*sigh*

one of the most important skills that is utterly drilled into you in academia is critical thinking. In an arts degree, you must learn to dissect an argument – if you cannot analyse effectively, you won't get through. In a science degree, you will be taught not only how to read papers, but how to critique them. Science is defined by the scientific method – and as tedious as it may sound, there is no skill more important than being able to scientific method your way through life.

The most important part of research is knowing how to do research – how do you find high-quality papers? What is high-quality evidence? Is it actually appropriate? What does 'significant' actually mean? The media loooves to deploy these terms without having a clue what the scientific meaning is – and as a result, the public get fooled. If you do not know how to do research, you will not find good information by doing your own research.

For example – members of the public who do their own 'research' into trans healthcare might come across the Cass Review, and see that it doesn't find much 'high-quality evidence' for gender-affirming care. They might think 'okay, guess trans healthcare is bad/experimental'. But if you've done academia, you'll know that 'high-quality' is just a synonym for 'randomised control trial' (RCT). For most research, RCTs are the gold standard. But RCTs require that people be unable to tell which condition they're in – that is, if they're receiving a placebo, they won't be able to tell. But with gender-affirming care, you can't exactly conceal the effects of hormones or surgery from a participant – after all, growing boobs/having a voice break/getting facial hair/altered sexual function is kinda hard to not notice. So, an RCT would actually be a completely inappropriate study design. Ergo ‘high-quality’ evidence is a misnomer, since it’d be really poor evidence for whether the treatment is helpful. As such, we know that so-called low-quality evidence is both more appropriate and perfectly sufficient. Academia is crucial to learning to debunk bad research.

And before y’all claim ‘but the papers being cited are all the ones that are getting funding – it’s just corrupt’ – you do realise being taught to spot bad science means that we get explicitly told how to spot corruption? Often, we can learn to tell when papers’ results are dodgy. Academia is about learning not to accept status or fame as a good enough reason to trust something. While I agree there are massive problems with some fields getting underfunded, or with science deprioritising certain social groups, abolishing science will only make this worse.

So. In conclusion, academia may be flawed, but it’s by far the best thing we have for learning to think critically. Anti-intellectualism is a blight on society, and horribly dangerous - it gets folks killed.