2 seconds in and already a better experience in terms of delivery and articulation compared to my current lecturer. Please continue to teach the wonders of statistics on youtube for the world's benefit.
in the R code when you add the value of count by one, if the candidate got accepted you increase the value of count by 1 and then you increase it again! since U add count by 1 outside of the if statement I think u should delete adding 1 to it inside if condition.
Thanks a lot for the video. Very precise and easy to understand. However, for choosing the value of scaling factor, its not always correct to choose the end value of support. I think its better to find out the local maxima. For example, taking the end value of support in a bell-curve would not be correct because we need to scale it to at least above the maxima of target function.
Thank you for your vid. I think i found a little mistake in your R-Code. Your counter will increased by two if your if-statement is true, which is, as far as i understood the method correctly, not what you wanted to archive.
you explained the concept behind the algorithm very clearly, it makes me easy to understand the whole thing, support! one thing I want to know is what is the usage in the reality? just curious
Simulating distributions. It's commonly used in software that models business processes. Not modeling them like Visio, though. Modeling them like, "how long does the average customer wait in line?" and, "what would happen if we added an additional cashier?".
Just one doubt. Generally we do not have the p.d.f, we just have some proportional density function. Therefore we just define a the eveloping cg(x) to be higher enough?
Pero lo que estás almacenando son las X, no los valores de pi(X) con ese algoritmo. Y las X son uniformes, no realizaciones de la función pi. Corrijanme si me equivoco.
2 seconds in and already a better experience in terms of delivery and articulation compared to my current lecturer. Please continue to teach the wonders of statistics on youtube for the world's benefit.
this is such a succinct video, broke down the method and explained how to apply it in such a helpful manner. Thank you.
This video is an act of kindness to me. Thanks for sharing
: )
This video is a masterpiece. Very well articulated.
in the R code when you add the value of count by one, if the candidate got accepted you increase the value of count by 1 and then you increase it again! since U add count by 1 outside of the if statement I think u should delete adding 1 to it inside if condition.
I am a python user, so not able to code R. However, this video is so very intuitive! Thank you for the nice lecture! Especially examples are so good!
I find your video really helpful and easy to understand thank you very much !.
thank u :)
So helpful!!! Thank you! You saved my midterm
these videos are gold. thank you so much
thanks for describe the theorem in the easiest way to understand. Best of Luck. the boss
Thank you so much for you clear explanations ! Really helpful
this is such an amazing video, subbing!
Wonderful video and well explained!
Thank you so much. Your video is really really helpful.
Thanks a lot for the video. Very precise and easy to understand. However, for choosing the value of scaling factor, its not always correct to choose the end value of support. I think its better to find out the local maxima. For example, taking the end value of support in a bell-curve would not be correct because we need to scale it to at least above the maxima of target function.
Thank you for your vid. I think i found a little mistake in your R-Code. Your counter will increased by two if your if-statement is true, which is, as far as i understood the method correctly, not what you wanted to archive.
This is really helpful, thank you!
you explained the concept behind the algorithm very clearly, it makes me easy to understand the whole thing, support! one thing I want to know is what is the usage in the reality? just curious
Simulating distributions. It's commonly used in software that models business processes. Not modeling them like Visio, though. Modeling them like, "how long does the average customer wait in line?" and, "what would happen if we added an additional cashier?".
Why do we need the second step u~Unif? In the third step, can we set the if condition as 1< Pi(Xi)/Cg(Xi) ?
Xi comes from a distribution that is hard to sample directly, which means that Monte Carlo methods can be used to approximate it
thank you , can you please tell me how can we do it in MATLAB
Very helpful vedios, but can you help me how to draw using this method from normal, exponential or any proposed distribution
How can we calculate the rejection ratio in this example?
This was an extremely good video
와씨 감사합니다 ㅠㅠ
Many thanks !!!:)
Thank you thank you thank you!
How does your pdf have values greater than 1?
where does numbers from count
am working on generated distribution, i want to simulate my true parameter using MLE pls help
Just one doubt. Generally we do not have the p.d.f, we just have some proportional density function. Therefore we just define a the eveloping cg(x) to be higher enough?
Yes, you can replace the target distribution pi(x) with a proportional distribution l(x) and the rest of the algorithm is the same
Great
Pero lo que estás almacenando son las X, no los valores de pi(X) con ese algoritmo. Y las X son uniformes, no realizaciones de la función pi. Corrijanme si me equivoco.
THX
amazing, but the R code is a little hard to read as it's a bit blurry. nevertheless, thanks so much
te amoooo :), pero creo q te falta un else