The better way to do statistics | Bayesian Statistics

Great video.
A little thing about the format:
-distractive images
-a bit hard to read the text and focus on what you say since the phrasing if often different
-The fonts with borders(contours) are hard to read.

I would like to know if there's ever been a drug approved where the control group had more deaths than the treatment group. But hey, Bayes.

This video is a work of art. Thank you very much!

Typo alert..I mean speak-o alert...on the outro, you said Bayes gave us this theorem in 1963 instead of 1763....

Bayesian statistics is based in subjectivity. Subjectivity in regard to human experience is unique and influential-able, and allows manipulation, that’s why corporations use it.

it should be pointed out if you used the 'informative prior' .. which is a delta function in the video then no matter what the data you wouldn't learn anything. you're prior has to be non-zero over the interval of possible values.

Hi, great video, I would appreciate if you would suggest a few reference books om these topics to study.
Thanks

“None of them were sufficient” , clearing throat
I see what you did there
😀

Last few months of my PhD. My ecological data is messy and made messier by stochastic events during my collection. Frequentist statistics have failed me. So now I find myself here...

Wonderful presentation, definitely better than I have been able to do so far. However, (maybe its somewhere down there in the comments). Bayes did it in 1763, not '1963.'

This video is fairly difficult because of the maintenance of jargon.

I got the sense that the Conjugate or the property of Conjugacy is when the prior distribution has the same shape as the posterior distribution i.e. the new observation doesn't change the prior distribution which means the chosen prior is stable despite the new evidence i.e. you've chosen a particular belief/set of parameters and they fit the data processed so far.

I may be wrong on this but that's the gist. Arbitrarily choosing a prior distribution is still a fanciful and mysterious thing given the information of this video.

I took a class on stochastic models which relied heavily on Bayesian methods. This video is helping me better understand my old notes. Thank you!!

majority of doctors fail simple questions on bayes theorem. hence the widespread misinterpretation of pcr testing that led to massively overstated case numbers

Interesting. As it turns out, my undergrad was heavier into Bayesian probability.

I have learned both ways in uni, it is interesting

I think in some ways the Bayesian method for testing vaccines make sense, it is a typical way to say the probability of getting covid given they have taken the vaccine is often used as a metric of how effective the vaccine instead of the probability of getting covid.

The probability of a 185 foot yacht named Bayseian sinking at anchor in some freak combination of circumstances ? I think some quantum entanglement double slit craziness had to happen .

I am thoroughly confused. "A" is treated both as the probability that you hit "Subscribe" and as the probability that you are a subscriber. These two events are not even approximately the same. In fact, they seem to be mutually exclusive. My brain fizzled. I may come back to this, but my hopes of getting some intuition were dashed.

Pretty baysed video.

Thanks!

1963 lmao

Being bayesian gives you the ability you conjure up a prior that perfectly matches the result you are trying to demonstrate, truly miraculous

The MCMC approach is one we used with STAN to generate models for the returns of financial timeseries (daily feeds). We had our Garch models setup and every parameter of each model had its own posterior distribution. The power with this was that we could a.) make forecasts continually every day and update it but more importantly, b.) detect changes across all the models when new data arrived. Further to this, being able to run scenario analysis was important and we used this during Covid to estimate recovery times for the assets (i.e. when to switch back into risky positions - out of cash into equities) and it worked really well. We didnt write the model to forecasts returns but rather it proved useful when forecasting volaility. It helps solve the approach of you fitting a model with fixed parameters and then updating it with a new set every 6 months. Instead, uncertainty was baked into the model via its parameters being distributions.

The benefit of the Bayesian approach was that we could account for uncertainty better and it worked with our risk management approach: Deploy, monitor, collect, update, test, deploy...etc.

🎉😅

Got lost half way thorough

@very normal, how good are Bayesian courses in brilliant?

Today I learned: some people pronounce Bayesian like “beige” rather than “Bayes”

I love bayesians.

Can we use the frequentist model to generate the prior knowledge if there is none?

Hooked me. Can you recommend some books that start from the basics of Bayesian statistics?

This almost feels like it could be used as a really transparent way to disclaim biases.

yeah, Nerd drama got me

In time label, fix the typo "Bayes' Theorem" from "Bas Theorem"

I am in love with your videos! What a treasure of information

Isn’t the bayes theorem the “gold standard” though?Since the bayes theorem is unknown. Thats why we use KNN or naive bayes?

Dude you're like a God I was understanding nothing about this and u helped me a lot thanks

its not a nerdwar lol, its 2 different methods for 2 different applications. Bayesian will be what almost all modern guessing is based upon. But there is a difference between guessing and documenting. "frequentist" is just the pure induction approach. Bayesian is where you apply logic on top of that , for application with limited data sets, or what I guess u could call dynamic systems.

U will have a hard time finding an actual statistician who thinks one is superior to the other, or thinks that pure induction is somehow more effective in finance, insurance etc.... I mean , its obviously not.

Blew my mind, to be honest. Never before have I considered how our expectstion of psrameter might impact conclusions themselves

Depending on the problem, which i choose to be the easiest way to solve will be the way.

Beysian Statistics is the only one I learned. I was in the computer curriculum though so the logical stuff is closely related.