Estimating the probability of a probability. My Patreon : www.patreon.com/user?u=49277905 Shoe icons created by Freepik - Flaticon www.flaticon.com/free-icons/shoe
Great video. Just started learning about Bayesian statistics, and we haven't been told why we were working with this distribution. This makes so much more sense now.
Your tuts are always my go-to resource when trying to learn some new Machine Learning concepts. Straightforward and beginner friendly, making it easy for me to understand. Thank you for your hard work.
Excellent explanation! I'm already somewhat familiar with the beta, but I really liked your treatment of the mean and variance and the relationship between those and the variance of the sampling distribution of a proportion. I thought that really it really tied the room together, like a really good rug.
Great video! It would be amazing to have the cards you are using available for purchase, or imagine an app where people can review topics using those cards that would be awesome! Very concise, yet triggers all the right questions.
Ritvik falling in love with Data Science thanks to you! You explain complex topics so well. Just curious which our your top 3 distributions in data science. Would love to hear the names. Thank you again for sharing your knowledge so generously!!!
Love to hear that and excited for your journey through data science. Hmmm right now my top 3 distributions would have to be: - Normal Distribution - Beta Distribution - Poisson Distribution But there’s so many other good ones out there!
So favorite distributions: beta, normal, and Poisson. Glad you did a video on beta distribution. I've found beta distribution to be really helpful in the real world. You can boil alot of things down into a binary outcome in the real world. Like if someone is going to click something or not or make a purchase. Something I wish most curriculum did was dive in more deeper into distributions. Alot of classes teach it but don't really go into detail on how you can apply it. Wish for example, there was more simulation assignments or fun real world applications of it. Thanks for helping bridge the education gap and making these videos fun and easy and informative.
Finally found the intro I've been looking for on this topic. Really clear in how you explain the concepts without getting bogged down in equations. Question: can this be used even if I'm trying to model very low probability? Eg after sixty days R=0 and B=60. Does mean and sd still make sense?
@ritvikmath Thanks for the great explanation! Have you done more on Bayesian stats and how to apply the prior binomial distribution to the beta? I'm been looking for explanations of Bayesian updating.
Thank you for the great video! I just have a quick question! So the empirical proportion becomes the mode right? But you later explained that the mean ends up being the empirical proportion. Shouldn't the two be different?
Was reading an article and came across the Beta distribution so wanted to understand more about it. Thank you for the clear and intuitive explanation. I think it's kinda funny that the distribution is written as Beta(Success+1, Failure +1). I mean, why not just let it be Beta(Success, Failure), it's simpler that way. Is it just stats being stats, or is there a good reason behind this?
I was buying a used book through Amazon this evening. Three resellers offered the book at essentially the same price. Here were their ratings: 94% positive out of 85,193 reviews 98% positive out of 20,785 reviews 99% positive out of 840 reviews Which reseller is likely to give the best service? Do you approach this question thru beta distribution?
