Видео

nice explanation

I was literally crying until I saw this video, thank you so much

How to interpret the data?

Kant asserts that the category of causation exists in the mind prior to observation? Then why is everything not caused? In other words does observation confirm causation; or does causation confirm observation: nothing that we observe is uncaused? There are no conscious perceptions that aren't caused, according to Kant, it's just that we have to correctly identify the causes? Observation only sees things that are caused but, at the same time, does not reveal those causes? Then what does confirm the cause? The mind? Isn't that the same position as Hume? No. So, according to Hume, observation hints at causation but it is the mind that imposes causation. Causation is really constant conjunction of observations being assumed by the mind as having necessary connection- causation. The mind imposes causation by assumption, not necessary connection. The mind imposes causation by short circuiting expectations and imposing necessary connection where none exists. Causation, according to Hume, is an illusion based on a lazy mind trying to save energy by imposing order on a chaotic Nature trying to kill it. Causation is a survival strategy of a being with limited energy and abilities. A necessary assumption for a being with limited means. Whereas, according to Kant, observations are entirely necessarily connected- caused. The mind's role isn't to impose causation but to expose the causal links. Instead of "assuming" causation the mind deduces causation by experiment in order to narrow down the causal relationships that most matter. Causation isn't an assumption of a lazy, evolutionary burdened, mind but a deduction of a rational, evolutionary burdened, mind. So, according to Hume, observation doesn't confirm causation. It can never confirm causation. It can only assume (impose) causation based on limited means trying to conserve energy. According to Kant observation does confirm causation, it cannot help but do so because it is inherent in everything that can be observed. However the causal "relationships" must be sussed out, and this is done by the mind through experimentation and reasoning. Instead of an assumption by a mind with limited means it is a deduction/induction by a mind with time for experiments.

Hume was an idiot and Kant should have remained in physics

thanks a lot

Crisis of absolutism. Hume was on his way to relativity thinking, a pre-Einstein if you will. In the world there are connections all the time, and knowledge is based on this. Our language describing Nature’s workings is the issue here. Hume’s theory of causality is bad logic. And doesn’t understand how our brain works. Once you abandon “necessary” and “guarantee” logic, we move forward. Just observe and describe then share and discuss the descriptions.

Great video, thank you very much , note to self(nts) watched all of it 9:48

I would recommend everyone to turn on CC

Seems like Kantian "schema" is a lot like Jungian archetype, then!

contiguity does not mean close to, it means touching

Just what I needed, thank you so much

great thanks

Great video! One question: Why don't you use the default Prior P(M)? Why do you use the uniform prior distribution instead?

Excuse me sir, is JASP available for free and does it run multilevel mediation analysis?

Thanks for the hard work you put into these videos. As someone who produce similar content, I know the burden.

Your explanation is clear, concise, direct. Just perfect or.... we may say "Jasp perfect".

Very illuminating and interesting. Many thanks for the series on causality. Admire how abstract concepts are explained in such simple and entertaining manner.Thanks again for sharing such valuable content

Very cool video. I am currently struggling with a mixed factorial ANOVA in JASP. When I run the ANOVA, the F statistics match my output from SPSS (so the data is entered correctly), but a VERY non-significant interaction (p = 358) keeps returning a bayes factor moderately in favor of the alternative hypothesis (3.05). Any help would be greatly appreciated.

you're a boss! thank you

I think this is great. Thank you 👍🏻

Well done. Thanks.

Great video

you deserve subscription.

There is a mistake. The prior for the alternative hypothesis is too diffuse. This is Lindley's paradox in reverse. The Bayes factor of 3 trillion (!) should be a red flag that something is terribly wrong. If one prior has much lower entropy than another (spike vs normal distribution), then it will almost always yield a lower likelihood unless you get lucky. You're guaranteed to reject the null.

sound quality is very poor

Thank you for this video, it is really interesting to see the mechanics of this. I know you are explaining how Jasp does this job, but can you elaborate on where P(data | M,θ) comes from? Also, is P(θ|M) equal to 1-cdf of a cauchy distribution centred on 0 with scale 0.707? I would like to be able to go through this "by hand" to fully understand how the analysis works. Would appreciate any help

Really good. Could you please post the data link?

What does it mean to say that there is a necessary connection between causes and their effects? Why does Hume think that we cannot justify our belief that there is a necessary connection between causes and their effects? Does Hume think that causality does not occur in reality, or does he just think that we cannot justify our belief in causality? Was Hume's skeptical point about induction really just that induction cannot guarantee certain knowledge? That seems like an obvious point, because no inductive inference that I make is guaranteed to be true. If a moving billiard ball collides with a stationary billiard ball, the stationary one is not guaranteed to move (it could be glued to the table, for example). Is that really all Hume was saying?

Mesej yang jelas, struktur yang jelas, mudah difahami, terima kasih

Hi, thanks for the video and congrats... but the video is incomplete... Am I Wrong?

people may think its explained in a confusing way but this video is faaaar better than any other on bayes hypothesis testing

plato philseophrs are the worst.

both have to have their books burned.

Great video, thank you!

What happend if your data do not follow a normal distribution? Should I choose Mann-Whitney? And thanks for your videos!

Thanks, Dr Guillermo Campitelli for these outstanding lectures

and then he vanished and stoppej making jideos

yooooooooooooo this shit slaps bh the way im an alchoholic and i need help and im in 95,000 dollars of debt to the IRS, i sold my pizza plave to a man what owns a small subset of 75 Subway Sandwhocj chains for 30,000$ and i want to know if thats enough to get the IRS off my back thanks and also where can i gat a acopy of alpha brotocol for the ninedo ds

Thank you for this, Mr. Campitelli

stop it!!!!!!! i hate philosophy!!!!!

Hi thank you vor the video!! I have a question. I see on tables related to bf10 interpretation, values fro 1 to 3 indicating moderate evidence for h1 and over 3 strong evidence for h1. Using jasp like this the value of bf10 can not be over 1.... I do not understand this point, can you help me please?

Thank-you!

Hi Guillermo, can I have your contact information, I need your help for a JASP course.

Hi I recently ran several Bayesian independent samples t tests using the informed prior vs the default prior. I understand why the same analyses using the informed prior gives a bigger Bayes factor compared to the default, but I didn’t expect the effect sizes to all be smaller (and credible intervals narrower) when using the informed prior. Does anybody know why this might be?

This is a very interesting topic and there are not too many alternatives on RUclips but the acoustics are really bad (echo) and all the pauses, ehms and started anew sentences make it difficult to follow. It seems like this could be massively improved at not too high cost.

I've been doing some research about Ronald Fisher and the video was so helpful and interesting! keep it up!

What about these variables Teacher quality causes students achievement

What's the key difference between traditional and Bayesian ANOVA conceptually?